University of Southern California Dissertations and Theses

Consensus Of Role Perceptions In A Welfare Planning Council

(USC Thesis Other)

Consensus Of Role Perceptions In A Welfare Planning Council

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content CONSENSUS OF ROLE PERCEPTIONS IN A
WELFARE PLANNING COUNCIL
by
Paul MV Berry
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Sociology)
January 1960
UNIVERSITY OF SOUTHERN CALIFORNIA
GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES 7, CALIFORNIA
T h is dissertation, w r it t e n by
Paul McCoy Berry
u n d e r the d ire c tio n o f /§-s D is s e rta tio n C o m
m ittee, a n d a p p ro v e d by a ll its m em bers, has
been presented to a n d accepted by the G ra d u a te
S chool, in p a r tia l f u lf illm e n t o f re q u ire m e n ts
f o r the degree of
D O C T O R O F P H I L O S O P H Y
Date January.. 15 60
DISSERTATION COMMITTEE
Chair ju
TABLE OF CONTENTS
Page
LIST OF TABLES................................... v
Chapter
I. INTRODUCTION ................................ 1
Statement of the Problem .................. 1
General hypotheses
Methodology of the Study ................ 4
Nominal Definitions of Tool-Concepts Used . 6
Value of the S t u d y ....................... 9
Organization of the Material of the
Dissertation ........................... 10
II. THE LITERATURE.............................. 12
Elements of "Role Theory" Utilized in This
Study................................... 10
"Role" theory as related to the
hypothe se s
Development of the Concept of consensus
in role study
Summary................................... 48
III. STRUCTURE AND FUNCTION OF THE CASE WORK
COUNCIL................................... 00
Formal Structure .................... 00
Membership and organization
Antecedents to the present Council
The Council in relation to the chest-
council hierarchy
Functioning of the Council................ 05
Formal star ernent of purpose
Flow and content of work
Summary................................... 61
IV. THE STUDY DESIGN............................ 63
Formulation of the Hypotheses.............. 63
The general hypotheses and role theory
The research hypotheses as derivations
from the general hypotheses
-ii
iii
Chapter
Construction of the Questionnaire ........
The role consensus battery
Questionnaire sections with intra-Council
focus
Questionnaire sections with extra-Council
focus
Pretest of the Questionnaire .............
Statistical Tools Utilized ...............
The use of nonparametric techniques for
data analysis
Statistical techniques used in the study
The Interview .............................
Responses to the Questionnaire ..........
Summary.............„ .....................
V. FINDINGS OF THE STUDY .......................
Role Consensus by Length of Council Tenure
Research hypothesis 1
Role Consensus by Role Judge Categories . .
Research hypothesis 2a
Research hypothesis 2b
Goodness of Role Perception 0 .............
Research hypothesis 3
Role clarity .............................
Research hypothesis 4a
Research hypothesis 4b
Role Involvement.....................
Research hypothesis 5a
Research hypothesis 5b
Research hypothesis 6a
Research hypothesis 6b
Research hypothesis 6c
Some Selected Variables Rej-examined in
Different Combinations .................
Research hypothesis 7a
Research hypothesis 7b
Research hypothesis 7c
Research hypothesis 7d
Research hypothesis 8
Conceptualizations of Community Organiza
tion as Possible Anchorages for Role
Perception .............................
Some Likely Frames of Reference Utilized
as Role Perception Anchorages as Revealed
in the Interviews .......................
Interviews with laymen
Interviews with social workers
Summary ....................................
Page
87
109
111
121
125
128
130
13 0
136
138
141
147
156
163
168
185
iv
Chapter Page
VI. CONCLUSIONS............................. 191
Some Interpretations and Implications of
the Study................ 191
Some Suggestions for Further Study .... 199
BIBLIOGRAPHY.................................... . 202
APPENDIX A .................................... 207
The Questionnaire
Cover Letter Sent with Each Questionnaire
APPENDIX B ................................... 227
Organizational Chart of the Greater St.
Paul Community Chest and Councils
Member Agencies and Organizations of the
Saint Paul Case Work Council
APPENDIX C ................................... 231
Statistical Tables Referred to in the
Dissertation
LIST OF TABLES
Table
1.
2.
3.
4 o
5.
6.
7.
Distribution of Questionnaire Respondents,
by Role Judge Groupings and Length of
Council Tenure . . „ .....................
Summary of the Probabilities Associated with
the Occurrence of Values Regarding Role
Consensus and Length of Council Tenure,
when Each of the Categorical Roles Is
Perceived by Role Judge Groupings . . . .
Rank Order Distribution of Potential Consen
sus Scores of Role Judge Groupings, as
They Perceive Each of the Categorical
Roles and Their Own (Personal) Roles . . .
Summary of Critical Values of Rho Depicting
the Degree of Consistency in the Percep
tion of All Categorical Roles, by Role
Judge Groupings .........................
Summary of Chi Square Critical Values and
Associated Probabilities Showing the
Extent of Independence of Stated Role
Clarity and Length of Council Tenure,
when Each Council Role Is Perceived by
All Delegates ...........................
Summary of Chi Square Critical Values and
Associated Probabilities Showing the
Extent of Independence between Stated
Role Clarity and Type of Role Judge
Grouping, when Each Council Role Is
Perceived by All Delegates ..............
Chi Square Table Showing the Extent of
Independence of Role Involvement Scores
and Types of Role Judge Groupings . . . .
Ranking of the Three Sources of Influence
Perceived as Being Most Important to the
Performance of Each Categorical Role, by
Role Judge Groupings .....................
Page
126
132
137
139
142
145
148
150
v
vi
Table Page
9. Ranking of the Three Sources of Influence
Perceived as Being Most Important to the
Performance of Each Categorical Role, by
Length of Council Tenure .................. lbb
10. Chi Square Table Showing the Extent of
Independence of Role Involvement Scores
and Stated Role Clarity, when the Lay
Role Is Perceived........................ 157
11. Summary of Chi Square Critical Values and
Associated Probabilities Showing the
Extent of Independence of Modal Consensus
Scores and Role Involvement Scores, when
Each of the Categorical Roles Is Perceived 160
12. Axial Orientation Scores of Role Judge
Groupings, as They Perceive Their Own
and the Categorical R o l e s ............... 167
TABLES IN APPENDIX C
1-Ci Percentage and Numerical Distribution of
Stated Role Clarity with Which Each Council
Role is Perceived by All Delegates, by
Length of Council Tenure .................. 232
2-C. Percentage and Numerical Distribution of
Stated Role Clarity with Which Each
Council Role Is Perceived by Each of the
Role Judge Groupings...................... 233
3-C. Chi Square Table Showing the Extent of
Independence of Stated Role Clarity and
Length of Council Tenure, for Each Council
Role........................................ 234
4-C. Chi Square Table Showing the Extent of
Independence of Stated Role Clarity and
Type of Role Judge Grouping, for Each
Council Role............................... 23b
b-C. Chi Square Table Showing the Extent of
Independence of Role Anchorage in
Professional Orientation and Role Judge
Grouping................................... 236
vii
Table Page
6-C. Chi Square Table Showing the Extent of
Independence of Role Anchorage in
Professional Orientation and Length of
Council Tenure ............................ 236
7-C. Chi Square Table Showing the Extent of
Independence of Role Involvement and Role
Consensus, for Each Categorical Role . . . 237
8-C. Chi Square Table Showing the Extent of
Independence of Role Involvement and
Stated Council Role Satisfaction for all
Council Delegates ........................ 238
9-C. Chi Square Table Showing the Extent of
Independence of Role Involvement and
Stated Council Role Interest for All
Council Delegates ........................ 239
10-C. Chi Square Table Showing the Extent of
Independence of Stated Role Clarity and
Role Consensus when the Categorical Roles
Are Perceived by All Delegates.......... 240
CHAPTER I
INTRODUCTION
Statement of the Problem
This study examines the role perceptions which each
delegate to a welfare planning council has of the role
which he himself plays in the council, and also of his es
timate of how the other council members, in this same or
ganization, perceive their respective roles. The council
contains three basic categories of roles, namely, Social
Worker, Layman, and Staff.
The organization within which the research was con
ducted is the St. Paul Case Work Council. It is an inte
gral part of the Greater St. Paul Community Chest and
Councils, Incorporated (Minnesota). (Hereafter it will be
referred to as the Council.) The organization functions as
a coordinating body for the public and private welfare
agencies which provide professional case work services for
persons in the St. Paul metropolitan area. It is composed
mostly of social workers, who are selected from among the
member-agency personnel, and of laymen, who serve on mem
ber agency boards of directors.^
■^The writer functioned as a lay delegate to the
The Council's "role" structure has been examined to
determine the amount of existent unanimity which character
izes the various role incumbents' perceptions of each of
the categorical roles that are contained within the Council.
The unit of study, therefore, is the individual's percep
tion of a given r o l e . 2 The main questions asked about
these perceptions are: are there differences among the var
ious groupings of role-perception judges; if so, how are
the differences related to certain important variables,
such as the length of time in the organization, profession
al training, role involvement, and selected role-perception
anchorages?
Council for six years prior to making this study, three of
which were served as Chairman. The organization is recog
nized locally as being highly productive in the welfare
planning field and its importance to the improvement and
expansion of case work services in the community is well
recognized in local we Ifare circles.
%or justification of the social psychologist's con
cern with perception, see Theodore M. Newcomb, "Social
Psychological Theory: Integrating Individual and Social
Approaches," Social Psychology at the Crossroads, ed. John
H. Rohrer and Muza £ e r Sheri~f (Ne w Yor k: Harper"& Brothers,
1951), pp. 48-49. "In summary, the two basic facts with
which the social psychologist must deal are those of the
organization of group life, and of the organization of
individual behavior. . . . It is the unique and supremely
difficult task of social psychology to provide an inte
grated theory of how human life goes on at both levels
simultaneously. . . . Interaction, particularly in the form
of taking and perceiving roles, is a necessary part of
process by which individual organization and social organ
ization simultaneously place place."
General Hypotheses
Tv'jo sets of hypotheses
study, both of which are shown
in general role theory. Likewi
supported by the writer’s exper
participant observer. One set,
serves to delimit the scope of
lating the study to the body of
other set is referred to as "Re
are specific in nature and are
Three inclusive ge
study.' (1) Laymen and so
in cornrnuni
spective c
Staf f ) dif
entially s
Social wor
patte
ce ive
socia
patte
the- r
time-
such
have been posited in this
in Chapter IV to be rooted
se, both are shown to be
ience in the Council as a
called "General Hypotheses"
the study and to aid in re
general "role theory." The
search Hypotheses," which
the ones to be tested,
hypotheses were used in this
orkers, when engaged jointly
ctivities , perceive the re-
ial Worker, Layman, and
the roles are not differ-
group structure.^ (2)
a professional educational
Social Work degree per-
neral
cial w
ty welfare planning a
ategorical roles (Soc
ferently, even though
pecified in the formal
kers who have followed
rn leading to the Master of
the respective categorical roles differently from
1 workers who have not followed such an educational
rn. (3) Both social workers and laymen will perceive
espective roles in a different manner as the length of
varies, during which they have been participating in
activities.
The direction of the difference is specified in the
Research Hypotheses, which are presented in Chapter 4.
4
The eight "Research Hypotheses," which are detailed
in Chapter IV, are concerned primarily with consensus of
role perceptions by different role-judge groupings in terms
of the following variables; length of time in the Council,
stated role clarity, anchorages of role perceptions, extent
of role involvement, and role satisfaction and interest.
Methodology of the Study
The primary instrument of research was a question
naire. It was designed to test the eight research hypoth
eses concerning role perceptions of the council delegates.
Three sets of role-cue statements were composed, embodying
the following factors: (1) council functions and policies,
(2) responsibility for initiatory action, and (3) simulated
quotations which the respondent was asked to attach to the
various role incumbents. Each delegate was asked to re
spond to each set of role-cue statements in terms of: (1)
his perception of them in terms of his own personal role;
(2) his estimate of the way other incumbents, in the same
categorical role as himself, would be likely to perceive
them; and (3) his estimate of how the role incumbents, in
each of the other two categorical roles, would be likely
to perceive them. Modal patterns, based upon the role per
ception data which the questionnaire yielded, were then
constructed for each categorical role. These modal patterns
became the major means for determining the degree of exist-
5
ent role consensus for each categorical role.
Two major categories of factors related to role per
ceptions were also explored by the questionnaire, namely:
(1) intra-council factors, such as council role involve
ment, and the degree of stated role clarity; and (2) extra
council factors, having to do with forces outside of the
Council which serve potentially as role perception anchor
ages, e.g., professional training and community relation
ships. An index of role involvement was constructed which
was based principally upon the amount and kinds of council
participation displayed by the delegates. The data which
this index yielded, was related to a number of other vari
ables, e.g., role consensus, role clarity, and role satis
faction.
Questionnaires were sent to all of the council dele
gates (127), and follow-up interviews were held with a
stratified sample of all delegates. The interviews were
pointed mainly in the direction of securing data concern
ing possible anchorages for role perceptions, as the latter
were revealed by the questionnaire data.
Non parametric statistical techniques were utilized
to determine the statistical significance of the data re
vealed by the questionnaire. These included the Sign Test,
the Spearman Rank Correlation Coefficient, and the Chi
Square test. The level of statistical significance was set
at the .05 level.^ Percentage distributions were utilized
1 = ^
to suggest possible areas for additional study.
Nomina 1 Definitions of Tool-Concepts Used
The following nominal definitions represent the ways
in which the major terms and concepts have been utilized in
this study.^ While some may be unique to the study, it is
believe that they are consistent with role theory and with
current social psychological usage.
Categorical role.— The behavioral expectations for
any one of the three groups of council participants (Social
Worker, Layman, and Staff), as reflected by responses to
questionnaire items.
Council.— The St. Paul Case Work Council (St. Paul,
Minnesota ).
Good role perceivers.— Those role judges whose role
perceptions approximate the modal pattern for the various
categorical roles.
“ ^Sidney Siegel, Nonparametric Statistics for the Be
havioral Sciences (New York: McGraw-Hill 6ooTc Company, Inc. ,
1956), p. 9. A more rigorous standard did not appear to be
necessary as important policy decisions would probably not
be made on the basis of the study findings.
^The rationale for this is discussed in Chapter IV.
^For a statement on "nominal definitions," see Paul
F. Lazarsfeld, "Problems in Methodology," Sociology Today,
ed. Robert K. Merton, Leonard Broom, and Leonard S. Cottre11
(New York: Basic Books, Inc., 1959), pp. 64-65.
7
Layman.--A delegate to the Council who is represent
ing some agency or an organization board (or the community-
at-large), and who is not a paid worker of that agency or
organizat ion.
Long-term participant.— A delegate to the Council
who, at the time of the study, was either serving his third
year or had served longer than three years.
Modal pattern.— A profile depicting the majority-
response configuration of expectations for a categorical
role, as set by a group of role judges, in response to a
set of role-cue statements.
Modal score.— A numerical value (derived from the
modal pattern) standing for the extent to which a given
role judge approximates the modal pattern, as set by all
role judges, in his role-judge grouping.
Role clarity.— The respondent’s perception of how
definite the role expectations are among incumbents of each
role judge grouping.
Role consensus.— The extent of unanimity (beyond a
majority) among the grouping of role judges concerning ex
pectations for one of the categorical roles.
Role-cue.— A normative statement submitted to all
council delegates for the expressed purpose of determining
whether or not it would come to be attached consistently
to any categorical role by role judges.
Role incumbent. — The occupant of a social position
in the Council, as represented by one of the categorical
roles.
Role interest.— The respondent’s perception of his
interest in council activities, as compared with his "most
interesting community activity."
Role involvement.— The extent of identification of
the council delegate with the Council, as measured by the
kinds and amounts of participation, as included in the in
dex of involvement.
Role judge.— A council member, as he responds to the
items of the questionnaire.
Role judge grouping.— A category of role judges.
Role perception.— (Used in three different ways.)
(1) a role judge’s view of his own personal role; (2) a
role judge’s estimate of how other incumbents of his cate-
^The term "grouping" is used instead of "category"
to reduce confusion with "categorical role." Each grouping
is based on different amounts and kinds of training, and
background of delegates. Two types of breakdowns are used:
(1) Social Workers and Laymen, and (2) Social Workers with
Master of Social Work degrees (MSW) and Social Workers with
out such degrees (Non-MSW).
gorical role are likely to view their own categorical role;
and (3) a role judge’s estimate of how the incumbents in
each of the other categorical roles are likely to view
their own roles.^
Role satisfaction.— The respondent's perception of
his satisfaction derived from council participation, as
compared with his "most satisfying community activity."
Short-term participant.— A delegate to the Council
who, at the time of the study, was either serving his sec
ond year or had served less than two years.
Social worker.— A paid worker, attached to a public
or private welfare agency, who is representing that agency
in the Council.
Staff.— The community organization specialist who
devotes full-time to the administrative functions of the
Council. (He is assigned to it and supervised by the Exec
utive Director of the Community Chest and Councils.)
Value of the Study
There are four ways in which the findings of this
study may be of value: (1) the light which may be thrown
upon group process and structure within an important organ-
O
Its use indicates the exact meaning intended.
10
ization in the life of a metropolitan community; (2) the
use to which the data can be put to facilitate the social
process of inter-communication between persons involved in
community welfare planning activities; (3) the implications
which the data may have for related organizations and agen
cies where the layman is brought into close working rela
tionship with trained specialists, e.g., education, city
planning, recreation, and health; and (4) the questions it
may help to answer as to the further research use of the
concepts of "perception," "role-cue," and "role consensus"
as tools for the study of role structure.
Organization of the Material
of the Dissertation
Selected elements within the general body of role
theory, which are basic to the design of this study, are
examined and interrelated with one another in Chapter II.
A summary description of the origin, formal struc
ture, and function of the Council is presented in Chapter
III.
The design of the study is detailed in Chapter IV.
Three major generalized hypotheses regarding role phenomena
in the Council are introduced first, together with their
justification from the body of .role theory, as presented in
Chapter II. The research hypotheses are then stated, and
the reasons for the assumption of their validity are given.
These reasons are shown to be rooted in role theory and
verified by the experience of the writer in the Council.
The various segments of the questionnaire are described in
detail, and the kinds of data which they are designed to
yield is shown. Chapter IV also describes the pretest and
the interview procedures used, and discusses the statistical
tools which were adopted for the study.
The results of the study are given in Chapter' V.
Here the findings from the tested hypotheses are presented
with the significance which is to be attached to these
f indings.
Chapter VI contains the conclusions drawn from the
findings of the study, as they are related to the body of
role theory presented in Chapter II. Some limitations of
the study are pointed out, with suggestions of how they
might be remedied in future studies. Suggestions are also
made for possible follow-up studies, particularly with
reference to data revealed which were not supported at the
level of significance set for this study.
CHAPTER II
THE LITERATURE
Although the term "role" has its roots deep in the
theatre, it has been widely used in sociological literature
since Cooley-*- and A/lead^ developed it in relation to the con
cept of "self." Park and Burgess, giving American Sociol
ogy its first comprehensive textbook, used "role" as a key
O
term in their definition of personality."-1 However, the use
of this concept of "role" has not been restricted to the
field of sociology; all of the behavioral sciences have
adopted it. Some social psychologists see it as having
high potential to bridge the gap between the psychologist’s
interest in human behavior at the individual level, and the
sociologist’s concern with the individual at the level of
■ T 4
social organization.
^Charles H. Cooley, Human Nature and the Social Order
(New York: Charles Scribner* s ' Sons", 1902), pp. l5T-T>S.
2g eorge H. Mead, Mind, Self and Society (Chicago:
University of Chicago PiTe'ss , 19^4 ), Part S .
^"Personality may then be defined as the sum and or
ganization of those traits which determine the role of the
individual in the group." Robert E. Park and Ernest W.
Burgess, Introduction to the Science of Sociology (Chicago:
University of Chicago Press, 19^1), pT ?0.
^Theodore M. Newcomb, "Social Psychological Theory:
Integrating Individual and Social Approaches," Social
12
13
Prior to the 1940* s, the concept was used largely as
an abstract generalization (often in a figure-of-speech
manner), with many different meanings and overtones. Some
of these uses represented heavy borrowings from the disci
plines of cultural anthropology and philosophy. It re
mained for those empirically oriented sociologists, social
psychologists, and group-dynamics specialists, who began to
"rediscover" the small group, to give the concept of "role1 '
its large present place in the behavioral sciences. They
demonstrated its utility for measuring group structure and
interpersonal interaction.
Something of the rapid, present-day advance in the
refinement of the concept can be seen in two contrasting
statements made only a few years apart by students of the
subject. Neiman and Hughes, writing in 1951, after having
examined the "role" literature to that date, conclude that
"The concept 'role' is at present still rather vague,
nebulous, and non-definitive."^ Yet Sarbin, writing just
three years later, sees the term "role" as having ". . .
recently come to have the beginnings of a stable core of
Psychology at the Crossroads, ed. John H. Rohrer and Muza-
fer Sherif (New York: Tlarper & Brothers, 1951), Chapter 2.
In the same book, see also: S. Stansfeld Sargent, "Concep
tions of Role and Ego in Contemporary Psychology," Chapter
15.
^Lionel J. Neiman and James W. Hughes, "The Problem
of the Concept of Role— A Resurvey of the Literature,"
Social Forces, XXX (December, 1951), pp. 141-49.
14
meaning . . ."6 To be sure, these statements rnay .reflect
different ways of looking at the matter; yet, on the other-
hand, in the light of the volume of empirical studies in
recent years, these statements may well reflect some ad
vance toward uniformity in the meaning and use of the con
cept. Perhaps what is still more significant is that this
is an outgrowth of empirical studies, rather than an impo
sition from some sector of social psychological "orthodox-
y."
There is no intent to imply here that there is
emerging a clear notion of what "role" is. Many defini
tions are still extant. But there seems to be a crystal
lization of some recognized approaches by which this aspect
of group and culture may profitably be studied, with a re
duction of some of the previous confusion. Perhaps even
more important, a body of "role" theory is emerging from
which to project more meaningful hypotheses, and into which
may be fitted certain findings gained from studies in many
different settings. With such a body of theory, it becomes
increasingly possible to grasp some relativity among the
many studies currently being reported in the social psy
chological literature.
^Theodore R. Sarbin, "Role Theory," Handbook in
Social Psychology, ed. Gardner Lindzay (Cambridge: Addison
WesTey Publishing Co., 1954), I, p. 224.
15
Elements of "Role Theory1 1
Utilized in This Study
It is not the purpose of this study either to trace
the development of the concept of social "role" or to de
fend its use as the basis for this study. The growing
volume of empirical literature on "role" furnishes ample
support for its use in this study. An able survey of the
development of the concept has already been made by Neiman
and Hughes.^ In this chapter, therefore, only such aspects
of "role" as are deemed relevant to the hypotheses for
this study will be discussed.
"Role" theory as related to
the hypotheses ~
Several major generalizations, implicit in "role"
theory, have been made in the formulation of the study
hypotheses. (1) Becoming a functioning member of a group,
and consequently being assigned a position In the group,
an individual views that position in terms of role norms
and expectations, and inter-role relationships. (2) In
creasing the length of time a person functions in a given
position, correspondingly increases the likelihood that
he will perceive the role of an other, as well as the self
role, more as other incumbents perceive these same roles.
^Neiman and Hughes, loc. cit.
16
(3) Having shared common experience prior to entry into a
group, group members tend toward homogeneity in the way in
which they view the various roles within a grou p. (4 ) Us
ing common reference-groups as anchorages for role percep
tions, group members perceive their own roles, and the roles
of others, in a homogeneous manner.
Does the body of "role" theory give any logical sup
port for these generalizations? To attempt to answer this
question, the literature in the field of "role theory," and
some of the findings of "role" research, were examined.
That groups have structure is well established in
sociological literature.^ one of the approaches to the
examination and measurement of these structures lies in the
concept of "social role" and the twin concept "social posi
tion. "
Social position. — If people are to function with
unity in the achievement of common goals, there must be
some order in their manner of interaction, so that the en
suing social process may be productive and satisfying to
the group members. They must be able to locate one another,
anticipate the behavior of each other, and have some mutu
ally accepted norms by which their own behavior in the group
kjVluzafer Sherif and Carolyn W. Sherif, An Outline of
Social Psychology (New York: Harper & Brothers^ ~ i ' 9 5 " £ > ),
pp. 1BI-233.
17
may be patterned.9
Persons who function within a group come to occupy
a social position, which is the location an acting indivi
dual, or category of acting individuals, occupies in a net
work or system of social relationships .-*•9 Further, a given
position does not exist as a separate entity, but only in
relationship to other positions in the group. Newcomb
points this out when he says, "Since every position is a
part of an inclusive system of position, no one position
has any meaning apart from the other positions to which it
is related.
Not only is a position viewed in relation to counter
positions within the group, but it must be seen in terms of
the total functions and goals for which the group exists.
This is another aspect of group structure. In an estab
lished group, positions are antecedent to the arrival of an
individual to the group. The continuity of position helps
to stabilize the group. This, in turn, makes group exper
ience satisfying to the individual. The advantage of such
stability is pointed out by Stogdill.
9Ibid., pp. 237-77.
l^Neal Gross, Ward S. Mason, and Alexander W.
McEachern, Explorations in Role Analysis (New York: John
Wiley & Sons, Inc., 1958J, p. 67.
■^Theodore M. Newcomb, Social Psychology (New York:
Dryden Press, 1950), p. 277.
18
A stable structure of positions has the effect of
eliminating the necessity of having the structure con
stantly under contest. The members are no longer re
quired to contend for position, a process involving un
pleasant tensions, but can devote their energies to
other purposes.
Social role.--Another aspect of group structure is
social position's twin concept, "social role." Behavioral
norms tend to gravitate toward, and form configurations
around, social positions. The concept of "role" is used to
refer to this aspect of group structure. "Role" is used in
the social psychological literature as referring to "a set
of behavioral expectations" which are applied to the in
cumbent of a given position.
There is no complete consensus among persons working
in the field as to what is to be included within the con
cept of "role."1^ There is, however, general recognition
that there may be a discrepancy between the behavioral ex
pectations for the role as viewed by the group members, and
1
the actual behavior of the persons who play the role.
l^Ralph M. Stogdill, Individual Behavior and Group
Achievement (New York: Oxford University Press, 1959),
p T ’ JT.-------
13Gr oss, Mason, and McEachern, op♦ cit., p. 67.
Coutu, "Role-playing vs. Role-taking," American
Sociological Review, XVI (April, 1951), pp. 180-87.
•^Daniel J. Levinson, "Role, Personality and Social
Structure in the Organizational Setting," Journal of Ab
normal and Social Psychology, LVIII (March, 1959), pp.
1TZ-7T. --------- ------- *-------
19
Newcomb deals with this recognized discrepancy with two
differentiating concepts: (1) the "role pattern" (role
expectations) and (2) the "role behavior" (actual perform
ance of the person).^
What function does a role perform for the group?
Newcomb specifies two very important ones; (1) a role pro
vides a pattern of expectations for the individual, and
channels his behavior as he comes to function in the posi
tion which he occupies in the group; and (2) it provides a
basis for communication between the interacting members of
the group.-^7 With reference to the latter function, New
comb points out that persons in the group come to be per
ceived in terms of positions and roles, and that these per
ceptions make it possible for common understandings to
exist between persons in the interacting situation. Being
crucial to the study, the first of the two above named
functions needs more extensive development. It involves
the matter of mutuality of behavioral expectations, which
is basic to group structure and to any situation in which
interaction takes place.
Kingsley Davis has an incisive statement regarding
role expectations.
The essence of any social situation lies in the
■ n i n i ^ »— m i i i M i — n w — w ■ a m m w T T B iiii^ iin m iii— n m T r i — r-— — mm— t t t — ------ M m u n m n r — -— i— — —
•^Newcomb, op. cit. , p. 459.
17Ibid., pp. 293-294.
20
mutual expectations of the participants. . . . Each
actor has some conception of what he himself expects of
others, and believes he has some notion of what they
expect of him and of what they expect that he expects
them to expect. . . . Assuming that everything "goes
smoothly" according to the mutual expectations, the
interaction usually achieves a previsioned end and gets
something accomplished.
Stogdill has developed a concept of "expectations"
which provides some cues as to how people come to join
groups, and also the manner in which the group influences
the individual's expectations.
The willingness of a member to permit a group to
structure his expectations is likely to be determined
by the extent to which he perceives the goals of the
group and the value systems of its members to be in ac
cord with his own value systems and those of the refer
ence groups which which he most strongly identifies
himself.19
Stogdill points out that identification of the individual
with the group purposes and norms comes quickly, under the
circumstances indicated above. He indicates, further, that
such an individual
. . . is likely to accept, or even to demand, structure
in a new group if he perceives a stable system of
goals, norms, and related expectations to reinforce his
own value systems and to increase the probability of
experiencing satisfying outcomes. However, the indivi
dual who expects to experience negatively valued out
comes and who perceives the goals of the group to be in
conflict with his own value systems may oppose the group
structure even though he accepts membership in it and
conforms sufficiently to retain his membership.20
•^Kingsley Davis, Human Society (New York: The Mac
millan Company, 1949), p. &4.
^Stogdill, og. cit., p. 76. 2Qlbid.. pp. 76-77.
21
What has the group's structuring of the individual's
expectations to do with social role? Stogdill develops
this relationship by suggesting three factors which operate
in the structuring of role expectations.
First, is the nature (status and function) of the
position he occupies. Second, is the demand made upon
him by the members as a result of changes in the struc
tural and operation requirements of the group. Third,
is the members' perceptions of the kind of person he
is. 21
As Stogdill develops the theme of role expectations fur
ther, he specifies that there are two aspects which must be
considered, namely, "responsibility" and "authority."22
He suggests that these two aspects define for the indivi
dual in the group the role that he is to play, and the re
lationship which his own role has with other roles in the
group. Persons whose role status is found in the upper por
tion of a rank system are expected to initiate action com
mensurate with the amount of responsibility attached to this
position. The necessary authority to accomplish this is
usually given such a person. Persons who occupy positions
somewhat lower, however, "may be held accountable for the
operations of subgroups," while persons located at the lower
end of the status system, according to Stogdill, "are held
accountable only for their own p e r f o r m a n c e s . "23
Stogdill gives a further cue to the behavior of the
2^Ibid. , p. 129. 22Ibid., p. 130. 23 Ibid.
22
individual (within the framework of role theory) who re
mains in the group but is not an active participant in it.
Some members may have overlapping roles. Others may
have roles that are lacking i.n definition. The bound
aries of roles are often as difficult to determine as
the boundaries of groups. Insofar as the boundaries
of a member*s role are vague and indeterminate, his
freedom of action is restricted because he does not
know how to act. This restriction may or may not be
imposed by other members. However, any testing behav
ior that the member exhibits in an effort to determine
boundaries may involve him in error, difficulty, repri
mand, or rebuff. Yet, the member whose role is lack
ing in definition is under compulsion either to seek
structure which will permit him to act according to the
expectation or to resign himself to inaction as a means
of avoiding the unpleasant consequences of possible
error.24
Before moving on to another major facet of role,
attention should be called to one supplementary aspect of
role expectation which is relevant to this study. It has
to do with the classification of types of group roles.
Sargent and Williamson, drawing upon R. L. Warren, suggest
three types, namely, "cultural,” "personal," and "situa-
tional." Although this classification does not entirely
fit the manner in which roles have been classified in this
study, it is valuable in that it suggests a place for a
kind of role which is not necessarily categorical in nature.
24Ibid. , p. 132.
25S. Stansfeld Sargent and Robert C. Williamson,
Social Psychology (New York: The Ronald Press Company,
1959), p. 3S4.
23
In commenting on their classification of "personal" r o l e , 26
these writers indicate that
. . . Personal roles are not in disagreement with
cultural standards but involve more individual varia
tion, as in the reciprocal relations between a partic
ular husband and wife or a given teacher and his stu
dents. 27
Up to this point, two major ideas have been dis
cussed, namely: (1) that two important elements of group
structure are "position" and "role," and (2) that the group
structure provides the individual with a basis for his ac
tions and his role expectations which is essential to the
interaction process. The third idea to be developed is
that of "taking the role of the other." This concept is
crucial, as it is implicit in the questionnaire which was
used for the study.
Taking the role of the other.— There are consider
able differences in point of view, as expressed in the role
literature, on the use of the concept of "taking the role
of the other," which will not be debated here. Because of
it relevance to the study design, Turner’s interpretation
of the concept will be adopted in this study.
Role-taking in its most general form is a process
of looking at or anticipating another’s behavior by
26jhis is introduced here since the questionnaire
used in this study asks the respondent to "perceive" in
terms of his "personal role."
2^sargent and Williamson, loc. cit.
24
viewing it in the context of a role imputed to that
other.28
This statement calls attention to a phase of role-perceiv
ing which is neither "playing-at" a role, nor is it simply
what one comes to accept as being appropriate to one1s own
role. Role-taking, for him, refers to an imaginative
process which enables one to put himself in the place of
the other. He further sees it as an integral part of role-
playing (role enactment), in that the individual must take
the role of the other before he can enact his own role
adequately. It may well be, Turner points out, that the
individual can be in error in this matter of taking the
role of the other, in that he may not mirror accurately the
real view of the other. This matter, however, is not ger
mane to the present study. The fact that this perceived
"role of the other" is likely to be within the total frame
of reference, in one's definition of the situation, is what
is significant here.
Turner takes the position that there are two possible
ways by which the individual may take the role of the other,
namely, by "projection," and by using it simply as "know
ledge."^ In the first case, "one constructs the other role
^Ralph H. Turner, "Role-taking, Role Standpoint and
Reference-group Behavior," American Journal of Sociology.
LXI (January, 1956), p. 316.
29Ibid., p. 318.
25
as he would if he himself were in the situation. The
"other” role is used as "knowledge," according to Turner,
when one bases a judgement upon prior experiences with a
particular individual, or with some other individual who is
presumed to be like him.
The entire concept of "role of the other" is devel
oped further by Turner by using two additional supplemen-
O I
tary concepts, namely, "standpoint" and "reflexiveness. "°-L
He utilizes the term "standpoint" to cover those situations
where the actor allows the inferred role of the other ac
tually to become the director of his own action. When one
takes the "standpoint" of the other, he acts as he thinks
that person would act. He may, on the other hand, be fully
aware of the role of the other, and this awareness may sim
ply constitute certain "knowledge" which is taken into ac
count in his own behavior. In the latter case, he is "tak
ing the role of the other" without allowing the other's
point of view to become his own. Instead of taking the
"standpoint" of the other, he is using his knowledge or
awareness of the other as a kind of "personalized third
O o
part or depersonalized norm." ^ By way of summary of the
concept of "standpoint," Turner states:
When role-taking includes adoption of the standpoint
of the other, the role-taking process is an automatic
■^^Loc. cit. ^^Loc. cit. ^Loc. cit.
26
determiner of behavior. One simply acts from the
standpoint of the role. When the standpoint of the
other is not adopted, some other factor must intervene
to determine the kind of influence which the role im
puted to the other will have on the actor.33
"Standpoint,1 1 as developed by Turner, is not an ap
pendage to role, but an integral part of it. In fact, he
makes it central to the whole concept of role, and sees it,
for example, as making possible both the manipulation of
others and the adjustment of persons to each other.
Turner's component of role-taking which he refers
to as "reflexiveness" has to do with the actor's own self
behavior, which he develops in much the same way that Mead
developed his concept of "self." Self notions reflect the
world of social experience, particularly as they are re
lated to the role of the other. Turner relates "reflex
iveness" to both "standpoint" and "knowledge," as described
above.
Turner’s analysis leads to the part which empathy
plays in the process of taking the role of the other. This
phase is not deemed necessary for the present study, as
perception, rather than empathy, is the principal concept
employed here. However, that portion of Turner's work, as
reviewed, is valuable, in that it points to the manner in
which the role-player may view the role of the other, which
is precisely what the questionnaire designed for this study
^Ibid. , p. 319.
27
asks him to do.
One further aspect of role-taking, which should re
ceive some attention at this juncture, is concerned with
the manner in which it is related to, and has been speci
fied in this study as, "role involvement." Newcomb ap
proaches this matter through the concept of "belongingness,"
which he sees as being essentially a predisposition to take
a role within a group. By pointing out that role-taking is
not just a matter of carrying out the expected behavior at
tached to a given role, but that it also involves being
responded to by other members of that group, Newcomb shows
the tie between "belongingness" and "role-taking."
Thus the attitude of belongingness includes the more
or less confident anticipation of being responded to by
others as they take their roles in understood ways.
Thus, we may define the individual's attitude of belong
ingness as a predisposition to find motive satisfac
tion by taking a member's role in anticipation that his
own understanding of the group norms is shared by other
members.34
It would seem reasonable to assume, therefore, that if the
person has a clear picture of what is required of him in a
given position, and if he has reasonable assurance that
others in the group regard his role as he himself does,
then the likelihood of his being motivated to participate
actively in the group (called "role involvement" in this
study) is enhanced.
^Theodore M. Newcomb, "Social Psychological Theory;
Integrating Individual and Social Approaches," Social
28
Perception.— Basic to the process of occupying a
position in a group, of conforming to the role expecta
tions which are attached to that position, and of taking
the role of the other, is the psychological process of
"perception." Although Thomas and Znaniecki (The Polish
Peasant in Europe and America, 1918) embodied much of what
is dealt with under this heading in their concept of the
"definition of the situation," it is still a relative new
comer to the social psychologist’s kit of tool concepts.
A perception is the property of the individual, and
is an intraorganismic response to stimuli involving, ac
cording to Newcomb, three kinds of psychological proces
ses: "omitting," "supplementing," and "structuring."^^
These three processes are all highly selective. In most
social situations, there are more things to be noticed
than a person can possibly take into account simultaneous
ly. Consequently, some selectivity is forced upon the in
dividual by the very nature of this limitation, and, con
versely, much has to be "left out" of the situation. This
is the process of "omitting." Newcomb defines "supplement
ing" as the process of "rounding out the picture" or "fill-
Psycholoqy at the Crossroads, ed. John H. Rohrer and
Muzafer Snerif (New York: Harper & Brothers, 1951), p. 43.
^Theodore M. Newcomb, Social Psychology (New York:
Dryden Press, 1950), pp. 91-92.
29
ing in the details.The process of "structuring," to
him, involves giving "integrated meaning to the situa
tion," when one "sizes it up" in a single manner.
In role theory, acts of perception, according to
Sarbin, are:
. . .organized against a cognative background of
role expectations. Role perception may be thought of
as a sequence of behaviors in which the perceptual
response is the first part of a social act: the (usu
ally) silent naming or locating the position of the
other (from observed actions or inferred qualities),
which serves to locate the position of the self. The
second part of the social act is the motoric response,
the role enactment, in which the actor performs ac
tions appropriate to his location of the positions of
self and other.37
Sarbin thus traces the case of inappropriate role enact
ment, on the part of the individual, back to possible
faulty role perception, which in turn results in a faulty
locating of both the self and the other.
Bruner has constructed a theoretical model of per
ception, in which he suggests that the process of percep
tion "involves an act of categorization. 1,33 By this he
means that after the organism has been stimulated by some
appropriate "input," the person responds by placing or
fitting the "input" into some class or "category" of things
^Ibid. t p. 92. 37Sarbin, ^oja. cit. , p. 229.
^Jerome S. Bruner, "Social Psychology and Percep
tion," Readings in Social Psychology, Eleanor E. Maccoby,
Theodore 114.' Newcomb, ancl Eugene L. Hartley, eds. (New York:
Henry Holt and Company, 1958), p. 92.
30
or events, which need not be elaborate or complicated. It
might be simply, for example, the category of "sound," or
"good," or "wonderful." He indicates that most of these
categories are learned, on the basis of past experiences in
the culture in general, and through language in particular.
Bruner claims the reliability of this idea in the following
statement:
Not only are the categories learned, but we learn to
estimate the likelihood that placement of an event into
a category on the basis of a few cues will be "accur
ate"— by which we mean, predictive in the sense that a
closer look will bear it out or that it will be consen-
sually validated when other perceivers come on the
scene or it will be confirmed by technological inspec
tion. 39
In commenting further on the perception model,
Bruner suggests that "categories of events with which we
become accustomed to dealing are organized into systems or
structures." These may be bound together by such factors
as time sequence, or simple association. He further sug
gests that the availability of these categories, into
which "inputs" come to be located, are determined by indi
vidual need and by interest-states possessed by the indi
vidual.
Certainly, it calls for no great stretch of the
imagination to see that Bruner's "categories" might well be
what have been called in this study, "categorical roles,"
^Ibid. , p . 93 .
31
which exist within a group structure, and which are recog
nized by the participants in the group.
Role-cues.— In his treatment of role theory, Sarbin
picks up the notion of "cue," which was included in Brun
er’s perception model, and reports that it has been used
in a research design for the examination of role struc
tures. 40
Sarbin introduces his discussion of role perception
by indicating that "the building blocks of role percep
tion are the overt acts of others." Everything which the
"other" does is grist for this mill. He suggests that the
perception of these observable acts or attributes of the
person:
. . . leads to predictable role behavior on the part
of the perceiver (providing the system of role expec
tations of the perceiver is known). Perceiving the
position of the other, then becomes a matter of attend-
int to and organizing cues.41
Bruner's idea of "cue" has been translated into re
search technique for the study of role perceptions by Sar
bin and Williams. This technique takes the form of a
single verbal statement, which respondents are asked to
attach to roles. Sarbin and Williams report that they re
corded 38 statements which had been made by different per
sons in different role situations, and asked different
4 0 s a rb in , c i t . , p. 230. ^ lltaxci.
32
groups of persons to indicate the age, sex, and occupation
of both the person doing the speaking, and the person to
whom the statement was addressed.4^ These respondents were
further asked to indicate the like resultant action from
the given sentence. Modal response patterns were then con
structed for each category of persons responding to the
role-cue statements. Sarbin and Williams report that their
study showed that "good role perceivers" (those responding
according to the modal pattern) tended to have more self
characteristics in common, while "poor role perceivers" had
fewer of these self-characteristics in common.
Another approach to the use of role-cues made by
Sarbin and Hardyck consisted of "postural cues" in the form
of "stick figures." Forty-two such figures were designed
which showed the human body in different posture states.43
These figures were then projected on a screen, by means of
lantern slides, and the respondents were asked to indicate
what they considered to be the appropriate "action, state,
or other quality that was suggested by the stick figure."
Sarbin, in reporting on the two above-mentioned studies,
concludes:
For our sample of undergraduates . . . we can con-
^sarbin, ibid. , p. 230. The technique calls for
the respondent to identify a given statement with a role.
For example, "all right, boys, get in there and fight."
43Ibid., p. 231.
33
elude that certain postures are perceived as associated
with certain acts or with certain qualities of a per
son.44
On the basis of Sarbin’s conclusions regarding studies
where role-cues were used as the research tool, it would
seem safe to infer that this technique might well be em
ployed to identify either roles or the attributes of per
sons .
Role-cues have been used in the present study to deter
mine the relative amounts of role consensus concerning the
categorical roles within the group being studied. Obvi
ously, these cues do not exhaust the total number of be
havior expectations of a given role, nor do they set the
outer limits of the role configuration, and they may even
bear little relationship to objective reality. However,
the present study is not concerned with determining the
content of the various roles, as such. It seems reasonable
to assume, on the basis of the work done by Sarbin and
others, that if the members of the group tend to attach a
given role-cue (statement) to a particular role in a con
sistent and statistically significant manner, reliable
claims can be made concerning the amount of consensus which
exists among the group members as to what particular role.
This matter of role consensus will be treated in the latter
part of this chapter.
44loc. cit.
34
In summary, when a person enters a group, he comes
to occupy a "position;" the position to be occupied has at
tached to it by the group members certain behavioral "ex
pectations" which have come to be known to students in the
field as "role;" the behavior of the person in the group
tends to be channeled and influenced by these "expecta
tions;" the stability of the group process is dependent
upon a "role structure;" the standpoint of the "other" is
basic to the perception which the person has of his own
role, and of the "role of the other;" the act of perceiving
a role is essentially that of organizing "cues," and locat
ing them within certain available "categories;" and the
roles, in the group, may be identified by the use of "cue"
statements. It remains to be shown what factors may come
into play in the perceptual structuring of role-cues. Two
major concepts will be developed to this end, namely,
"reference-group" and "frame of reference."
Perception as influenced by reference groups.— Role
perceptions are not "free-floating" for they have anchor
ages. A substantial body of theory is developing around
the concept of "reference-group," which seems to show con
siderable potential for throwing light on this subject of
anchorages. Reference-groups are characterized by Sherif
as, "those groups to which the individual relates himself
as a part or to which he aspires to relate himself psychol-
35
ogically.
The concept of "reference-group" was developed, to a
very large extent, from the interpretations of the studies
on the American soldier in World War I I . I n this study,
the concept of "relative deprivation" was developed to help
explain some of a soldier*s responses to situations which
seemed, on the surface at least, to be void of both logic
or consistency.^^ It came to be understood that the degree
of deprivation which the soldier felt was, in part, a
function of the group which he had in mind when he made the
judgment. For example, when the soldier in a combat situ
ation compared himself with those "at home," he perceived
himself as being "bad off." However, if he shifted his
frame of reference to that group of persons who were in
more dangerous circumstances, then he tended to perceive
himself as being relatively "well off." Merton classified
the frames of reference used by soldiers as, first, those
4^Muzafer Sherif, "The Concept of Reference Groups
in Human Relations," Group Relations at the Crossroads, ed.
Muzafer Sherif and M. O. Wilson TNew S^or k : 'Harper & feroth-
ers, 1953), p. 206.
^Samuel A. Stauffer, e_t. aj.. , The American Soldier:
Adjustment Durinq Army Life (Princeton: Princeton Univer
sity T£ess,' m ^ T T V o l T T ; for subsequent development of
"reference-groups," see Robert K. Merton and Paul F. Lazars-
feld (ed.), Continuities in Social Research— Studies in the
Scope and Method ol "The American Soldier," (Glencoe: The
Free Press, 1950).
^Stauffer, ibid., p. 52.
36
attitudes of judgement which were influenced by comparison
with the situation of others with whom they were in actual
association; second, those judgements which came from com
parison with men having something of the same status as
themselves; and third, those judgements arising from com
parisons with men "who are in some pertinent respect of
different status or in a different social category," e.g.,
48
enlisted men compared with officers.
That people are influenced by the groups of which
they are a part is nothing new in social psychological
theory. However, what is relatively new, and what the
study of The American Soldier has so well demonstrated, is
that people are also influenced by groups to which they
have never been attached, and even by some which may exist
only in the imagination of the person. Reference-group
theory seeks to systematize the perceptual processes which
lead to the incorporation of the "psychological group"
within the judgemental frame of reference of the individual.
At this point, the question might well be raised:
how does the reference-group influence the perception of
"roles," or the psychological process of "taking the role
of the other?" According to Kelley, reference-groups per
form two functions for the individual, namely, "normative"
^Robert K. Merton, Social Theory and Social
Structure (Glencoe: The Free Press, 1957), p. 231.
37
and "comparative. "49 pje sees the former function as exer
cising control over the behavior of the individual. It is
the latter of these two functions, however, the "compara
tive," with which the present study is primarily con
cerned, where the individual is seen to have "psychologi
cal membership" in the group, even though he may have never
become a part of it. This "psychological group" is used as
the basis for comparisons, and in making perceptual judge
ments. Its values, standards, or behavioral norms consti
tute a point of anchorage for subsequent evaluations of the
"self" or of the "other."
To be consistent with the earlier discussion of
"role-taking," it should be pointed out that one need not
necessarily adopt the "standpoint" of the reference group,
but, rather, that "reference group" may come to be used as
the "third party." Turner suggests the alternative manner
in which the reference group may serve, in regard to the
concept of "standpoint," with the following statement:
So long as the actor is using the reference group
only as a point of comparison in estimating his own
social standing or in deciding whether to be satisfied
or dissatisfied with his lot, external attributes of
the other alone are involved. The role of the relevant
other is not being taken. But when levels of aspira
tion, degrees of determinations, and the like are being
AQ
Harold H. Kelley, "Attitudes and Judgements as
Influenced by Reference Groups," Readings in Social Psy—
chology, ed. Guy E. Swanson, Theodore M. Newcomb, and
lugene L. Hartley, (New York: Henry Holt and Company, 1953),
pp. 412-13.
38
compared, the individual must necessarily tgke the role
of the other in order to make a comparison. 0
In view of the fact that some persons belong to, or
aspire to be accepted by, many groups, what factors seem to
be important in the selection of the reference group to be
used in a given situation? To answer this question, New
comb suggests:
The degree to which a person’s membership group
serves also as a reference group depends upon the de
gree to which his membership in it brings him satis
faction or dissatisfaction.
Newcomb states further that a group may serve as a negative
reference group.
For a thoroughly dissatisfied or disgruntled group
member, . . . the unwelcome membership group may serve
as an equally potent reference group, but in reverse.
It may be said to serve as a negative reference group,
in the sense that the individuals attitudes are in
fluenced by a set of norms which he shares to the point
of understanding them . . . but not to the point of ac
cepting them.^2
Thus, it may be seen that the "positive" reference-
group is one in which the person is motivated to be ac
cepted and treated as a member, while the "negative" refer-
ence-group is one in which he has come to be motivated to
reject, or oppose, or possibly just to .ignore.
In reporting on the work of Centers and of Steiner
(unpublished manuscripts, separate studies), both of whom
^Turner, ^0£. cit. , p. 327.
^Newcomb, op. cit. , p. 226. ^ i o c . cit.
39
worked with the influence of reference-groups in the area
of social class, Newcomb concludes:
This study (Centers) indicates clearly that certain
attitudes of large numbers of people are better under
stood in terms of their positive reference groups than
merely in terms of their occupational membership
groups . . .53
In reporting on Steiner, he states that
. . . at least part of the experiment shows con
vincingly that the group which serves the individual as
a frame of reference does influence his attitudes.54
In the concluding statement to a report of his much-
quoted Bennington Study, which had to do with attitudes as
they were influenced by reference-groups, Newcomb contends
that
. . . in a community characterized by certain ap
proved attitudes, the individual's attitude develop
ment is a function of the way in which he relates him
self both to the total membership group and to one or
more reference g r o u p s .5b
Although many of the above citations, which support
the importance of the concept of "reference-groups, 1 1 came
from studies related to attitude formation and change, it
seems reasonable to assume that the concept has relevance
to the study of perception as well. Attention will now be
directed toward the concept of "consensus," as it relates
^Newcomb, ibid. , p. 230. ^ Tbi.d. , p. 232.
^5rheodore M. Newcomb, "Attitude Development as a
Function of Reference Groups: The Bennington Study," Read
ings in Social Psychology, ed. Guy E. Swanson, Theodor’ s M.
Newcomb, and-Eugene L. Hartley (New York: Henry Holt and
Company, 1953), p. 430.
40
to the degree of unanimity which exists among the members
of a group, concerning the expectations of a given role.
Development of the concept of
consensus in role study
Much of the building of role theory, and some of
the research in the field, has been based upon the assum
ption of role consensus. To examine this assumption, one
must look at some of the still more basic assumptions
which undergird the concept of culture, as it has been
utilized to help explain human behavior.
Consensus as an implicit generalization. — Cultural
anthropologists, in search of the sources of patterned
behavior, have tended to assume the unimodality of behav
ior patterns that relate to societal positions and sets of
functions. To be sure, in more recent years, they have
recognized the difference between "real culture" and the
average "behavior of persons" as not being necessarily the
same. Yet, underlying this, has been more or less the
tacit acceptance of the unimodality of the culture pattern
itself. What has brought this generalization into question
is the fact of deviant behavior, which has been perceived
more as a function of personality differences than of cer
tain cultural factors.
Linton, in attempting to delineate the varieties of
role behavior to be found in connection with a given role,
41
proposed the notion of norms as ranges of acceptable behav
ior rather than as fixed points.^ However, many of his
readers tended to overlook this notion in favor of a larger
concept for which Linton was also responsible. His concept
of "cultural constructs" was, in essence, modal behaviors
abstracted from an infinite series of potential varia—
tions.^7 Gross, Mason, and McEachern, who have examined
what they call "the postulate of role consensus" more
critically than most of the current role empiricists, have
this to say about Linton*s approach to the matter:
Linton, then was certainly aware, as are most an
thropologist s , that in any given situation the observer
will probably find a range of behaviors. He apparently
assumed, however, that the distribution of behaviors
for a specific situation in a society usually has a
single mode, and this assumed mode he defined as a cul
ture construct. For many situations this assumption of
unimodality is undoubtedly reasonable, but to maintain,
as Linton apparently does, that most real culture pat
terns are unimodal is to state a tenuous proposition.
It minimizes the importance of the possibility that
for some situations there may be bimodal, trimodal,
multimodal behavioral distributions.^8
Gross, Mason, and McEachern push the matter further
by pointing out that what may go for deviant behavior may
actually be more a matter of lack of consensus around an
^Ralph Linton, The Cultural Background of Person
ality (New York: Appleton-Cientury £o'.7 1945), p. 4b.
^^Ibid., pp. 45-46.
^Neal Gross, Ward S. Mason, and Alexander W.
McEachern, Explorations in Role Analysis: Studies of the
School Super intendency kole I New York: j'ohn Wiley and S7»ns,
incTv i w ; p . 25:
42
assumed mode, and hence not actually deviant behavior at
all.
When Linton defined social role, he likewise ap
peared to assume unimodality and consensus. In one of his
definitive statements, for example, social role is char
acterized as being:
. . . the sum total of the culture patterns associ
ated with a particular status. It thus includes the
attitudes, values, and behavior ascribed by the society
to any and all persons occupying this status.59
Gross, Mason, and McEachern insist that the postu
late of consensus is imbedded in Lintonfs phrase, "ascribed
by society," and that the "untested" assumption is that the
members of a society hold the same expectations for incum
bents of the same position.
This assumption of unimodality of role expectations
apparently is not limited to certain cultural anthropolo
gists. Social psychologists, too, appear to have operated
for the most part from this same base. On the other hand,
Newcomb, to whom we are greatly indebted for his ability to
conceptualize the various facets of role, seems not always
to have been clear in his thinking about the matter. Some
times he seems to assume unimodality and, at other times, to
be cognizant that such may not be the case. The former may
^Linton, j0£. cit. , p. 77.
^Gross, Mason, and McEachern, ^ojo. cit. , p. 30.
43
well be inferred, when he says:
Both behavior standards and norms for perceiving
people are shared by all members of any group, but they
apply in distinctive ways to different members of the
group depending on how these members are classified.®1
This statement certainly seems to assume consensus. Here
differences in role behavior would seem to be accounted for
in terms of the sanction available for their enforcement.
However, in other statements, it is obvious that Newcomb
sees the need for recognition of the fact that all persons
may not see a given role alike. For example, with refer
ence to carrying out the functions attached to a given
social position, he suggests three different categories of
role expectations: (1) those necessary, (2) those permitted,
and (3) those prohibited— all residing along a continuum.^2
He states that not all behaviors will be performed by every
occupant of a position, and thereby gives room for disagree
ment between role occupants as to what is necessary, for
example, but such is not explicitly stated. The crux of
this distinction seems to continue to rest upon the degrees
of importance of certain behaviors vis-a-vis other behav
iors .
It should be pointed out, however, that later on
Newcomb sharpens the matter somewhat when talking about the
role of the mother, for he states:
^Newcomb, ^og. cit. , p. 276. ^Ibid., p. 281.
44
A criterion of 50 per cent agreement that any given
behavior is demanded of ail mothers might be set up.
In that case the prescribed role would consist of all
behaviors which were considered by at least half of all
the respondents to be demanded of ail mothers.63
Sherif seems to regard consensus in a more rigid
manner than does Newcomb. Using the term "status'1 rather
than ’ position" to refer to the relationship of one indivi
dual to other individuals in the society, he insists that
once a person occupies a given behavior status, then the
corresponding pattern for behavior is fixed. As he says,
"Once he is there, in a particular status, he has no choice,
but fulfills the requirements demanded of the status."64
Sherif suggests no possibility of differences in perception
as to what might be expected in the way of behavior.
Consensus treated as a variable.--Gradually the
notion of role consensus is seen as an independent vari
able, beginning to work its way into the empirical studies
on role. Some examples have been selected as illustrative
of the increasing importance attached to this dimension of
social role.
Cottrell appears to be among the first to take this
dimension seriously. In a study reported by him in 1942
63Ibid., p. 282.
^Muzafer Sherif and Carolyn W. Sherif, An Outline
of Social Psychology (New York: Harper and Brothers, i^bo)
pTIBT ------
45
(working with the adjustment of persons to age and sex
roles, in terms of the clarity with which the roles are
defined), he hypothesized that "the degree of adjustment
to roles which a society assigns to the age-sex categor
ies varies directly with the clarity with which the roles
are defined."^ According to Cottrell, one of the more
important factors which might influence this matter of
clarity is "inconsistency in the expectation exhibited to
the individual by members of his social world.Here we
have expressed one of the earliest recognitions in empiri
cal literature that consensus, or lack of it, may throw
some light upon role behavior.
This same theme of role clarity is picked up by War
ren in a study reported in 1949, concerning the relation
ship between social disorganization and social role. War
ren hypothesized that, "Social Disorganization varies in
versely with the clarity of definiteness of cultural
roles."67 in commenting on this hypothesis, and on his own
study, Warren says:
Conflicting conceptions arise as to the behavior
^Leonard g. Cottrell, "The Adjustment of the In
dividual to His Age and Sex Roles," American Socioloqical
Review. VII (October, 1942), p. 618.---------------- ----
66Ibid., p. 619.
^Roland L. Warren, "Social Disorganization and the
Interrelationship of Cultural Roles," American Socioloqical
Review XIV (February, 1949), p. 84.
46
appropriate to a cultural role, and while a sequence of
behavior may be demanded of some, that very sequence
may be discouraged by others.68
Concerning consensus, Warren has this further to say:
What is considered a gross mismanagement of office
by the wider public may be but the meeting of expecta
tions of loyalty to a family or closely knit political
group.
It can be seen, then, that lack of consensus ex
presses itself in terms of conflicting conceptions of
cultural roles by different groups.
In 1964, working with leader roles in the U. S. Air
Force, Davis raised the question of the extent of agree
ment between persons in official leader positions and those
in follower positions, as to how the leader should perform
his duties, and the degree of agreement associated with
follower adjustment."^0
Hall, in examining the relationship between amounts
of "cohesiveness" among members of an aircraft crew and
role consensus, makes this relationship a subject for con
sideration in two of his four major hypotheses as follows:
Hypothesis 1: The greater the cohesiveness of the
crew the greater the agreement among crew members on
role prescriptions for the aircraft commander.
Hypothesis 2: The greater the resultant force on
any individual to remain in the crew the smaller the
discrepancy between his role prescription and the
crew's mean role prescription for the aircraft com-
68Ibid. 60Ibid.
^ F. James Davis, "Conceptions of Official Leaders
Roles in the Air Force." Social Forces. XXXII (March, 1954),
p. 253-58.
47
mander.
Finally, one additional illustrative study will be
cited which should serve to turn attention to a study-set
ting somewhat similar to the one in which this study was
conducted. Thomas examined the matter of consensus in re
lationship to the quality of work performed by social
workers in different size organizational units.7^ He also
considered another dimension which he called "Breadth of
Role Conception," meaning the "number of activities or
functions conceived as part of the role .H e found that
the greatest role consensus seemed to exist between worker
and supervisor in the smaller units. "Breadth of Role
Conception" varied with age and experience, as well as
with the size of the work units. Commenting on his own
findings, Thomas states, "The fact that role consensus was
greater in the smaller bureaus may indicate greater cohe
sion of the primary groups and readier acceptance of the
goal to provide service.
Other illustrative studies, where consensus has been
considered a variable, have been made in a variety of so
7lRobert L. Hall, "Social Influences on the Aircraft
Commander*s Role," American Socioloqical Review, XX (June.
1955), p. 293.
7^Edwin J. Thomas, "Role Conceptions and Organiza
tional Size," American Socioloqical Review, XIV (February,
1959), pp. 30-TT
73Ibid., p. 32. 74Ibid., p. 37.
48
cial settings, such as marriage roles, occupational roles,
and bureaucratic roles. However, the studies mentioned
should establish the claim that role consensus has come to
be accepted as a justifiable variable for social psycholo
gical study.
Summary
Selected elements within the general body of role
theory, which are basic and germane to the design and hy
potheses for this study, have been examined and interre
lated. This process has brought into focus certain con
cepts which will be utilized, in the present study.
It has been pointed out that when the person enters
a group, he comes to occupy a "position." This position
usually has a set of behavioral expectations attached to
it, which are referred to as "role." Group structure, as
expressed in "position," "role," and "expectations," in
fluence the role perceptions of the group member. The
social psychological process of "taking the role of the
other" is essential to the process of role perceptions—
both "self" and "other." The "standpoint" of the "other"
is an integral part of the self role and of the role of the
other. "Reference-groups" serve as anchorages for role
perceptions. Role perceptions may be detected through the
use of "role-cues." The degree of "consensus" concerning a
given role by a grouping of role judges can be determined
through the use of "role—cues." "Role consensus," though
treated in the early role studies as a constant, has more
recently been accepted, among students of the field, as a
variable which can and should be measured.
CHAPTER III
STRUCTURE AND FUNCTION OF THE
CASE WORK COUNCIL
Formal Structure
Membership and organization
The St. Paul Case Work Council, the organization
used for this study, is one of three coordinating councils,
all of which are integral parts of the Greater St. Paul
Community Chest and Council, Incorporated, hereafter re
ferred to as Community Chest and Councils. The other two
are the Health Council and the Leisure Time Council. (See
Organizational Chart, Appendix B, p. 228.) Each of these
councils is composed of lay and professional delegates from
the member agencies and organizations which are operating
within the metropolitan area of St. Paul. Some of these
are state-wide in their coverage (St. Paul being the capital
of the state), and some are county-wide (Ramsey County), but
the majority are city-wide in scope and coverage.
The major type of service rendered by a respective
agency or organization determines in which of the three
coordinating councils it will have its primary membership.
(Some maintain membership in all three.) Each council has
an Executive Secretary, who devotes full time to the work of
50
51
his particular group, and whose salary is paid by, and who
is supervised by, the Executive Director of the Community
Chest and Councils. (The Executive Secretary will be
referred to hereafter as Staff.)
The St. Paul Case Work Council, hereafter referred
to as Council, is composed of those agencies or organiza
tions interested in and furnishing what is known in social
work circles, as case work services. Its membership ros
ter includes public agencies which receive their support
from taxation, private agencies which receive part of all
of their support from the Community Chest, and other inde
pendent agencies and organizations. (See Membership List,
Appendix B, p. 229. )
Each member agency or organization designates a min
imum of two official delegates. One of these delegates
is usually from the agency's paid staff and the other is
from its advisory or governing board.^ In addition to
these delegates from member agencies, the Council itself
elects not less than three or more than fifteen delegates
from the community at large. These must be voted on each
year and may serve for only three consecutive years.
The Chairman, elected annually, may not be on the
payroll of a public or private social agency, and conse-
^-Agencies with several different departments or
branches send two delegates for each distinct department,
e.g., the County Welfare Department has representatives for
Public Assistance, Child Welfare, etc.
52
quently is always a layman. He may serve for only three
consecutive years. Three Vice-Chairmen are constitution
ally provided for, and each, in contrast to the lay chair
man, must be paid employees of a welfare agency. One of
these must be from a private agency affiliated with the
Community Chest, another from a governmental or public
agency so affiliated, and the third may be from any affil
iated or participating agency.
The above officers, together with three additional
elected persons, constitute an Executive Committee which
functions as a steering committee for the Council. It
prepares recommendations on policy, supervises all commit
tee work, and initiates suggestions for studies.
Both the Executive Committee and the Council are
required to meet monthly (except July and August), the
latter having approximately sixty delegates present at each
regular meeting. The Executive Secretary serves not only
the Council as a whole, but also all committees within the
jurisdiction. This means that no committee or delegation
ever meets or functions without benefit of the professional
services of Staff.
Antecedents to the present council
The present council idea and structure has its roots
deep in the community chest movement which flourished sub
sequent to World War I. The relative success of the "War
53
Chest" and "Liberty Loan" drives had demonstrated the
effectiveness of united action in raising funds.
St. Paul, which was faced annually with some fifty
separate campaigns for fund raising to support its commun
ity services, organized its Community Chest in 1920, with
forty-one member agencies. But the Community Chest struc
ture did not limit the function of this new organization
to fund-raising and budget control. It moved in the direc
tion of a structure for the social planning of welfare
services by making provision for a Welfare Council, as is
demonstrated in the following statement;
Money raising as an end in itself has never been the
function of our Chest. Our founders made that clear
by setting up a Welfare Council, staffed by Chest per
sonnel, to follow closely the needs of the individual
agencies, and to study ever-changing social problems
and conditions in the c o m m u n i t y .2
However, an examination of the structure of that
early Welfare Council reveals that its functions were
linked closely to the administrative functions of the Com
munity Chest, with only a limited charge to "create special
committees to initiate and conduct surveys and study prob
lems . "2
In 1946, machinery was set in motion to reorganize
2st. Paul Community Chest, The Saint Paul Community
Chest in Its Twenty-Fifth Year, May^ 194b.
^St. Paul Community Chest, The Saint Paul Community
Chest After Fifteen Years, May, 1935.
54
the Greater St. Paul Community Chest and Councils, Incor
porated. This effort resulted in the structure which, for
the most part, is the pattern in operation today.
The Council in relation to the
che st-'c ou'nc'i 1 h ier a'r chy
Before examining the flow or operational procedure
of the Council, it may be helpful to see it in relation to
the other units of the chest-council structure. (See Or
ganizational Chart, Appendix B, p. 228.) It will be re
called that the Case Work Council is but one of three coor
dinating councils, all of which are identical in structure
and equal in power and privilege, representing different
aspects of the welfare planning operation at the lower
level of policy determination.
On the Organization Chart, the next highest body is
shown as the Planning and Research Council. This Council
is composed of twenty-five members who function at various
levels of the entire chest and council structure; some are
designated by the Board of Directors and some are repre
sentatives from one of the three coordinating councils.
This brings together agency representatives, persons re
sponsible for the Community Chest as a corporation, and
persons who raise the money to support the various serv
ices. This would appear to have high potential for making
the planning process realistic and responsible and for com
municating the needs and thinking of the member-agencies
55
directly to the policy-makers and money-raisers. The Plan
ning and Research Council is expected to coordinate the
work-products of the three coordinating councils, and may
accept or reject recommendations coming from them. It
determines whether the requests or recommendations (e.g.,
of the Case Work Council) are in line with the policy of
the total structure, whether they are feasible in the light
of available resources, and whether they are internally
consistent with what the other councils may be doing.
At the top of the structural hierarchy is the Board
of Directors of the Community Chest and Councils, which is
the final authority on policy and money matters. The Plan
ning and Research Council elects nine of the twenty-eight
Board members, and the chairmen of the three coordinating
councils are all ex-officio members of the Board of Direc
tors.
Functioning of the Council
Formal statement of purpose
Planning and coordinating welfare services are the
major stated functions for which the Council was called
into being and for which it now exists. These functions,
explicitly stated in the Council’s Statement of Organiza
tion, Article II (1948), and reaffirmed in a "Council Self-
evaluation Study" made in 1954, are as follows:
1. To coordinate case work activities of all
56
agencies in the St. Paul area.
2. To promote and develop a practical, consistent
and comprehensive policy of case work ser
vices.
3. To survey situations in particular areas to
determine case work needs.
4. To interpret the work of case work agencies to
the public, both in terms of the quality and
scope of service being rendered, and by
pointing to the necessity for extended service
as community need may justify.
5. To create public sentiment for the establishment
and support of necessary agencies and facili
ties.
6. To support desirable social welfare legislation.
7. To work in conjunction with the Planning and
Research Council and under its direction.4
Flow and content of work
To understand the interaction process as it takes
place within the Council, one must have some picture of the
content and flow of its work.
Source of work assignments.— At the formal level, it
can be said that the Council has complete control over the
choice of projects with which it is to be concerned. Re
quests for its attention and judgments may and do come from
a number of sources: the Planning and Research Council; a
member agency, either directly or through its delegates; a
delegate, who may, on his own, initiate action from the
floor at a regular meeting; the Executive Secretary, speak-
either as such or for the Executive Director of the Commun-
4Greater St. Paul Community Chest and Council, Inc.,
Statement of Organization of the Greater St. Paul Case Work
&ounciiT' January, l£4b, p. 1.
57
ity Chest and Councils; or from some group or agency
wholely outside the structure of Chest and Councils. (The
latter two usually come via the Executive Committee.)
Work content in one selected year.— In order to
furnish a rounded picture of the work of the Council, a
list of projects which moved through it during a single
year, 1954, will be presented.^
1. An evaluation of the program of a nursery school
operated by one of the settlement houses (at its own re
quest ).
2. The finding of areas of neglect with respect to
child care services— a study in connection with the grant
ing of federal funds for public agencies engaged in child
care.
3. A study of boarding home rates, for the purpose
of establishing uniform rates to be paid by both public and
private agencies.
4. A review of the effect of jurisdictional curtail
ment on services rendered by a given infant home.
5. A study of the need for an institution special
izing in treatment of the seriously emotionally disturbed
^Minutes of the Committee for Self-evaluation of The
Case Work Council, March 16, 1954 (in the Council files).
The year 1954 is selected because it was within the writerfs
term as chairman and hence he has greater familiarity with
the program.
child.
6. A study of some welfare bills being submitted
to the state legislature.
7. A study of the problem of transiency, and pro
grams related thereto.
8. Reconsideration of a policy established in 1946
connected with the Refugee Resettlement Program, sponsored
by one of the family service agencies.
9. Examination of a proposed long-term research
project aimed at developing types of classifications of
the symptoms and causes of family breakdown (subsequently
known as the Family Centered Project and financed by foun
dation grants ).
10. A review of case work services of an agency
serving mostly immigrants in the light of the total case
work picture in St. Paul, leading to possible inter-agency
agreements and coordination.
These are the types of work which flow through the
Council (insofar as a sample of one year’s activities can
portray), and which constitute the subject matter of the
interaction process which takes place within it.
Scope of Council action. How does the Council
handle such matters as were catalogued above? Most of its
work centers around study, the results of which become em
bodied in subsequent policy or program recommendations.
59
Taking direct action about some community concern would
perhaps be one exception to the general rule. For exam
ple, although it would not lobby for specific legislation,
it might take a stand with reference to some pending legis
lation and attempt to communicate that stand to those
groups or individuals supporting or lobbying for the legis
lation. ^ jt might send a representative to talk with the
mayor of the city about a given matter, but would not wage
a campaign for what is ordinarily known as social action.
The organization does a good deal of groundwork necessary
to inter-agency relationships or basic to intra-agency
changes in policy or program, but it is in no position to
enforce its judgments or recommendations. When it sets up
a committee to examine some aspect of an agency program, it
does so in an advisory capacity and always with the consent
of— and usually at the request of— the agency involved.
After the Council accepts a proposal for a study or
inquiry to be made, it either sets up a committee or refers
the matter to a standing committee. Usually, if the mat
ter is to involve much time, this committee will make peri
odic progress reports to the parent body. The Council’s
Executive Committee is supposed to be kept informed of the
^Not every delegate is empowered by his constituent
agency or organization to vote on certain public issues,
particularly those involving proposed legislation. This
fact obviously limits the social action function of the
Council.
60
approach and progress of each project through Staff. The
Executive Director of Chests and Councils keeps a detailed
flow chart of each project as it passes through each of the
coordinating councils.
When the Council receives the final report of the
committee, it may choose among several kinds of action:
accept it in whole or in part; refer it back to the com
mittee for further study or with recommendations; or reject
it. If it is accepted, the next usual step is for the
Chairman of the Council to present the report subsequently
to the Planning and Research Council, where it is viewed in
the light of the entire Chest and Councils program. Imple
mentation of the report, particularly if it involves money,
is largely determined at this level. As previously stated,
the Board of Directors of the Community Chest and Council
is the final authority on all matters.
The Council as information-dispenserThe Council
is more than a work body. The regular monthly meetings
usually provide information for the delegates concerning
various programs in operation in the community, or a speak
er may be brought in to present some aspect of community
need. For example, when the Housing and Redevelopment
Authority was engaged in a major slum clearance project,
its officer in charge of the relocation of families pre
sented some of the problem aspects of this process as they
were related to case work agency programs, and clarified
for the delegates the kinds of limits on the resources
available to the Authority. When the Council of Churches
engaged a social-work-trained chaplain to work with the
probation office, he was given an opportunity to explain
the jurisdiction of his functions, the referral process,
and his relationship to welfare agencies in the community.
An experimental project, which provided for a liaison
social worker who had previously been placed in the Police
Department for the purpose of detecting family disorgani
zation, was explained and discussed at a regular Council
meeting. Consequently, this organization functioned as a
clearing house for information for its member agencies
through their delegates. The delegates, in turn, are ex
pected to convey this information to their fellow workers,
their governing boards, and to citizens generally whenever
the opportunity affords itself.
Summary
The principal function of the St. Paul Case Work
Council is to facilitate and coordinate case work serv
ices. It is composed of lay and professional delegates who
represent all welfare agencies of the metropolitan area,
both public and private, and a number of organizations in
terested in what such agencies are doing.
The Council, which meets monthly, is one of three
62
coordinating councils, and is a functional part of the
Community Chest and Councils structure. Most of its work
consists of committee studies of existing programs or poli
cies, and the exploration of unmet welfare needs in the
total community. Its recommendations are often the begin
ning of policy and program change in welfare services, par
ticularly those involving inter-agency relationships.
This organization also provides a means for dissem
inating information about existing programs, and for the
presentation of welfare problem situations in need of at
tention. The Council both initiates its own projects and
accepts referrals from other units within the Chest and
Council structure and from outside groups.
CHAPTER IV
THE STUDY DESIGN
Formulation of the Hypotheses
The general hypotheses, which serve to set the
boundaries within which this study has been conducted, were
introduced in Chapter I. A review of the role literature,
which was deemed relevant to this study, was given in Chap
ter II. The task assigned to the present chapter is three
fold, namely: (1) the building of a bridge, so to speak,
from the body of role theory to the general hypotheses for
the study; (2) the presenting of the research hypotheses,
as they have been derived from the general hypotheses; and
(3) the describing of the tools and techniques used, and
the procedural steps taken to test the research hypotheses,
together with the rationale for each. These aspects of the
assigned task will now be undertaken in the order outlined.
The general hypotheses and
role theory
General hypothesis 1. The first general hypothesis
is that laymen and social workers, when engaged jointly in
community welfare planning activities, perceive the respec
tive categorical roles (Social Worker, Layman, and Staff)
differently, even though the roles are not differentially
63
64
specified in the formal group structure.^-
This hypothesis is based upon the existence of an
established group structure, which is antecedent to the
arrival of the person into the group, and which is inferred
from the behavior of the persons in the group situation.
Two essential elements of group structure, as outlined in
Chapter II, are "position” and "role." In the present
study, the focus of attention is on the latter. In this
first hypothesis, the generalization is being made that when
a person seeks to function in a group, he doea not do this
in random fashion, but rather his functioning is shaped,
channeled, and patterned by the expectations of others in
the group (see sections on Position and Role, pp. 16-23).
A further generalization is being made— which role theory
also would support— that these expectations tend to nucleate
around the positions and roles which are recognized and ac
cepted by the group members (see p. 18). In the case of
the Council, the three roles that are officially recognized
are that of the Social Worker delegate, the Layman delegate,
and the professional community organization expert (Staff),
who performs the administrative functions necessary to the
Council.
Chapter II also pointed out that the individual
^The direction of the differential is specified in
the Research Hypotheses.
65
functions in the group, and the self must be positionally
and normatively related to others in the group. To do this,
it is necessary to "take the role of the other" (see sec
tion on "Taking the role of the other," pp. 23-27), which
means that one must have some perception of that "other" in
the group.
The structure of a group may be characterized oper
ationally as being of either a formal or informal nature.
A formal group might have constitutional provisions, writ
ten job specifications, or a manual of operations. By
informal group structure is meant the shared expectations
which are accepted and sanctioned in the group process out
side of the formal structure. This informal structure may
be expressed in expectations communicated in day-to-day
contacts and in the exchange of ideas, during such times as
the coffee-break, the telephone conversation about a client
referral, and the small talk which precedes the formal
Council meeting. Sarbin uses the very expressive term,
"incidental learning," to cover these informal means where
by the role expectations come to be known and understood
and mutually reinforced.^ As there are no initiation cere
monies or orientation sessions for newly appointed or
elected delegates, most of the role expectations must be
■^Theodore R. Sarbin, "Role Theory," Handbook in
Social Psychology, ed. Gardner Lindzay (Cambridge: Addison
Wesley Publishing Co., 1954), I, p. 226.
66
gained through these informal and incidental methods.
In the Council, there is a minimum of formal recog
nition given to two of the three categorical roles: Layman
and Social Worker. Apart from the fact that the Chairman
of the Council must be a layman, and two of the three Vice-
Chairmen may not be laymen, there is little else specified
in the way of role differentiation in the formal structure.
However, the experience of the writer in the Council leads
him to believe that all of the roles are perceived differ
entially by the council members, regardless of the lack of
role specificity in the formal structure.
At the outset of this study, one unique aspect of
council roles had to be accepted, which appeared to be dif
ferent from the situation faced by students of such recip
rocal roles as manager-worker, superintendent-school board
member, or leader-follower. In these latter cases, many of
the role tasks are performed "solo" and often apart from
each other. These separately performed role tasks may thus
serve as the core of the role-expectation pattern for a
given role. When this is the case, role content analysis
^What has been said regarding the lack of structur
ing of the Layman and Social Worker roles does not hold
true for the Staff role. This role is given continuity and
specificity through supervision within the Chest's and
Council's hierarchical structures, through written job des
cription, and through recorded precedent in the form of
flow charts for projects undertaken, and through staff
meetings. In this Council, it is the only role so struc
tured.
67
is made relatively easy. However, in the case of the or
ganization studied here, all members contribute to the
decision-making process in a more or less interchangeable
fashion (referred to in the hypotheses as ’'engaged joint
ly"). Consequently, in making a role analysis in the
Council, it is more a matter, for example, of who might be
expected to initiate a given action in which all will
eventually take part, than of someone doing a set of tasks
which makes it possible for someone else to perform another
set of tasks; or it is a matter of how consistently certain
opinions, with reference to various proposals or plans,
might be expected to be held by certain categories of role
players.
Experience would seem to indicate that there is an
awareness on the part of both laymen and professional work
ers in the Council that they perceive their respective
"roles" differently, even though they "engage jointly" in
common tasks. For example, in the naming of a study com
mittee, there is usually an expressed awareness that it
should contain both laymen and social workers. Also, social
workers go to great lengths to attempt to interpret their
remarks or to translate their formal reports into layman
language. Laymen are proud of the products of the Council,
but often their remarks indicate they know that the bulk of
the content of the Council's recommendations is furnished
by the social workers.
68
The layman presumably comes to the Council with the
interests of the community-at-large in mind, and also to
look out for the interests of the agency which he formally
represents in the Council. Further, he also represents the
many contributors to the Community Chest, as well as the
taxpayer.
In brief, this hypothesis states that although the
delegates function jointly on common and undifferentiated
tasks, they perceive their respective roles differently.
General hypothesis 2. A second general hypothesis
is that social workers who have followed a professional ed
ucational pattern leading to the Master of Social Work
degree perceive their respective categorical roles differ
ently from those who have not followed such an educational
pattern.^
This hypothesis is concerned with what might reason
ably be expected regarding the behavior of persons in the
group who share a commonality of background before enter
ing the group, as well as with the continuity and lasting
behavior-determining effects of a long period of training
and indoctrination on the person, particularly when that
training is institutionalized to the point of being recog
nized as a major "port of entry" into a profession.
^See Research Hypothesis 2b for the hypothesized
direction of the differential.
69
The choosing of a profession by an individual usual
ly constitutes a long-time (often life-time) commitment,
and consequently self concepts are very much a part of this
process. Moreover, joining a profession involves not only
self-commitment but group acceptance as well. Being ac
cepted into a professional fraternity provides high motiva
tion for the individual to perceive professional situations
as others in the fraternity perceive them. Becker and
Carper point this out: "Participants in work institutions
tend to see themselves in relation to those upon whom their
success in these institutions depends.
This same idea has been explored by Merton, Reader,
and Kendall in their studies on the sociology of medical
education.^ They found that perceptions which medical stu
dents have of their role as doctor, was closely linked to
that degree to which they had come to become identified as
doctor by the patients. To enter a profession, one must
come to be accepted by the other practitioners as well as
the patients or clients— one must have the appropriate
credentials. Time spent in formal training helps guarantee
^Howard S. Becker and James Carper, "The Elements
of Identification with an Occupation," American Sociologi
cal Review, XXI (June, 1956), p. 345.
^Robert K. Merton, George G. Reader, and Patricia L.
Kendall, The Student Physician: Introductory Studies in the
Sociology of Medical Education t&mbrlcige: Harvard Univer
sity Press, 195Vj, p.
70
such acceptance,^ because it tends to contribute to homo
geneity of perception in at least two ways: (1) through
differential association during the training period; and
(2) through increased involvement in the professional role,
with increased investment of time and energy in it.
Not only are trainees in social work interacting
selectively over a long period of time, but they are ex
pected eventually to come to act and think like social
workers should. If they do this, to the satisfaction of
the instructors and supervisors in the training program,
they can be reasonably sure of being accepted into the pro
fession. Certainly, the training process, when coupled
with the prospect of the reward of a place in the profes
sion, would be expected to create some homogeneity of per
ception among social workers-in-training, concerning role
perceptions in welfare-related activities.
Professional education, as a potential conditioner
of role perceptions, might be viewed from another vantage
point. Essential to specialization in any field are: (1)
^See Chester I. Bernard, "Functions of Status Sys
tems in Formal Organizations," Human Relations in Adminis
tration, ed. Dale Yoder (New York! P'rentTce-Rall, Inc., ~~
Mb'lir P» 261. Bernard shows how status systems provide
orderly ways by which one can justify to one’s self, and
to the group, the manner in which he has spent past time.
Further, he discusses the personal need which the indivi
dual has, "... for the endorsement of the individual’s
past history as a creditable element in his existing per
sonality. "
71
the literature, and (2) an established and accepted manner
by which the recruit to the specialty is introduced into
the in-group. Graduate training for a specialty helps to
guarantee that the recruit has been introduced to a body of
accepted practice and theory as contained in the literature
and as transmitted by oral tradition.
This is true in social work, which today is highly
specialized, one of the specialties within it being social
case work. Most graduate schools of social work include
some formal training in the area of community organization
as part of the education of the case worker. Professional
social work training has a great deal to say about the
social worker-layman relationship. In like manner, the
case worker is made to be aware of the importance of inter
agency relationships, and is trained specifically in what
is called the "referral process." Being skilled in this
latter process calls for considerable knowledge about the
total welfare resources of the community, and how they com
plement and supplement each other. The graduate school of
social work, then, provides the trained case worker with
some common expectations on such matters as these. Conse
quently, the persons who come to the Council with this kind
of background can be expected to be more homogeneous in
their perceptions of Council functions than those who have
not shared these training experiences.
One problem, however, which arose in this study and
72
which called for something of an arbitrary decision, was
the need to decide where this kind of common background
could be said to exist with high probability. Some gradu
ate schools of social work give a professional degree known
as the Master of Social Work, while other schools give the
more traditional master's (academic) degree. For purposes
of this study, a line was drawn at the point of the Master
of Social Work degree, thus obviating the necessity of hav
ing to evaluate the various kinds of social work degrees,
as to the commonality of their educational content. This
means that some of the delegates in the non-master's group
ing of role judges, although they do not have MSW degrees,
have had graduate training perhaps identical with those in
the Master of Social Work group. Yet, those who have been
included in the MSW group, in all probability, can be ex
pected to have had a very similar kind of professional
training, and consequently represent, for purposes of this
study, a relatively high degree of homogeneity, insofar as
background training is concerned. The non-MSW group, on
the other hand, is an operational grouping, including per
sons who fall into the social work category generally, but
among whom there is a wide range of professional back
ground— from no professional training at all to, perhaps, a
good deal.
General hypothesis 3. The third general hypothesis
73
states that both social workers and laymen will perceive
the respective roles differently insofar as the length of
time during which they have been participating in such ac
tivities varies.
Two major variables are involved in this hypothesis.
The first involves an extension of the one treated under
the second hypothesis, concerning the matter of profession
al training; and the second is the variable of time.
Part of the discussion in the second hypothesis re
ferred to the professionally trained social worker versus
the untrained one, and this same comparative evaluation can
be said to exist to an even greater degree between the MSW
and the layman. The layman’s background and training ob
viously is not the same as that of the social worker.
There are various professional and occupational commitments
involved in the layman grouping of role judges. Some of
them have had training for business, some for law, while
others represent a variety of different kinds of profes
sional training. Some of the delegates have had no profes
sional training at all, e.g., some of the housewives. Of
this latter group, however, some are college graduates with
majors in fields related to social work. Yet the major
factor is that all of these people come to the Council as
laymen, regardless of the heterogeneity of their formal
education and training.
The second major variable incorporated within this
74
hypothesis is that of time. The longer the person is in
the group, the greater his potential for increased contact
and interaction with others, and the greater the potential
effects of "incidental learning" upon the person. Sherif’s
studies on the development of group norms gives support to
the belief that it takes time for the individual to gravi
tate toward the norm which has been set by the group.®
Gross, Mason, and McEachern, in their studies of the school
superintendent, concur in this theory.
The longer the members of the social system have
interacted with one another, the more consensus they
will have on the expectations they apply to incumbents
of positions in that social system.9
They indicate that this hypothesis was supported at the .05
level of significance in their study. Additional agreement
is found in an article by Reicken and Homans, when they
write about consensus and the length of time within the
group, for they say, "The degree of consensus may depend on
the members’ similarity in cultural and social background,
and also on the length of time the members have been in
interaction with one another.
^Muzafer Sherif, "Group Influences upon the Forma
tion of Norms and Attitudes," Readings in Social Psychology,
ed. Theodore M. Newcomb and Eugene L. hartley INew York:
Henry Holt and Company, 1947), pp. 77-90.
^Neal Gross, Ward S. Mason, and Alexander W.
McEachern, Explorations in Role Analysis (New York: John
Wiley & Sons, inc., l9b&;, p. ±77.
■^Henry W. Riecken and George C. Homans, "Psycholog-
75
Merton, too, in his treatment of role behavior with
in the bureaucratic structure, points out that depersonal
ization of the perception of clients appears to accompany
the prolonged occupancy of a bureaucratic position.H
Newcomb gives recognition to the time factor in his discus
sion of what constitutes a group. He believes that two
important elements are necessary, namely, shared norms and
common interests. The sharing of norms, he says, increases
with increased contact and interaction among the group mem
bers, and he recognizes this in the following statement:
New members of most groups do not usually accept
the group norms as fully as do older members. In fact,
one of the jobs that old members must undertake is to
imbue new members with the standards of behavior which
are thought to be appropriate to the g r o u p . 12
Thus, it would seem reasonable to assume that those
who had served on the Council for several years would be
more likely to perceive their functions, responsibilities,
duties, policies, and relationship to Council roles in a
more homogeneous manner than would the novices. Moreover,
the time element would seem to be even more important here,
ical Aspects of Social Structure," Handbook in Social Psy
chology, ed. Gardner Lindzay (Cambridge: Addison Wesley
Publishing Co., 1954), I, p. 788.
■^Robert K. Merton, Social Theory and Social Struc
ture (Glencoe: The Free Press, 19bv;, p. z b ' 2 .
12Theodore M. Newcomb, Social Psychology (New York:
Dryden Press, 1950), p. 493.
76
when we recall that the Council has no formal training
process for the newcomer.
Homogeneity in both role enactment and role percep
tion, then, would seem to be related to the amount of time
which the individual had spent in acquiring the role expec
tations and requirements. This hypothesis simply posits
this factor of time as one of the variables in the role
perception process.
The research hypotheses as
derivations' £rom the general
hypotheses
While the general hypotheses discussed above serve
both to help set the boundaries of the present study and to
act as vehicles to aid in bringing the body of role theory
to bear upon the study, the Research Hypotheses are de
signed for the stated purpose of being tested empirically.
The terms used in these hypotheses have been defined in
Chapter I.
Research hypothesis 1. Role consensus is greater
among long-term participants than among short-term partici
pants.
Role consensus, as defined in Chapter I, refers to
the extent of unanimity (beyond a majority) among a group
ing of role judges concerning expectations for one of the
categorical roles. Although the choice of how much agree
ment constitutes consensus is, obviously, an arbitrary one,
77
the usual meaning is majority. Newcomb, for example, sug
gests the use of the fifty per cent criteria in a discus
sion of the means for determining the role perceptions of
the mother role.^ Further, the size of each of the three
role judge groups did not seem to justify the search for
smaller consensus clusters within each grouping. For pur
poses of this study, therefore, consensus of role percep
tion is said to exist among a group of role judges only if
more than half of the respondents make the same judgment.^
Role judges, as specified in the discussion of the general
hypotheses, refers to one of three categories of Council
delegates: MSW, Non-MSW, or Lay.
A long-term participant (as defined in Chapter I) is
a Council delegate who, at the time of the study, was either
serving his third year, or had served longer than three
years (3+). A short-term participant is a delegate who, at
the time of the study, was serving his second year, or had
served less than two years (2-).
This first research hypothesis is simply a specific
expression of general hypothesis 3.
•^Newcomb, _o£. cit. , p. 282.
■^There is one exception to this. It is concerned
with determining the "majority" when three alternative
choices are open to the respondent instead of two. This
is treated in a subsequent section entitled, "Determining
the Modal Pattern."
78
Research hypothesis 2. (a) Role consensus is great
er among social workers than among laymen; and (b) role
consensus is greater among Master of Social Work degree
participants than among those without such a degree.
A social worker is defined, for purposes of this
study, as a paid social agency worker attached to either a
public or private welfare agency or organization, and who
is representing that agency in the Council. A layman is
defined as a delegate to the Council representing some
agency or organization board, and who is not a paid worker
of that agency or organization. The Master of Social Work
participant was defined in the discussion of general hypoth
esis 2 above.
In this hypothesis, the generalization is made that
social workers, by virtue of their commonality of back
ground and experience, would be more likely to be homogen
eous in their perceptions of Council roles than laymen who
have no such common base of experience and background.
This may be construed from the facts that social workers
(1) are engaged in professional tasks that are related,
(2) share a common professional literature and social sta
tus, and (3) have relatively frequent inter-personal con
tacts with each other. Further, the generalization is
being made (research hypothesis 2b) that the MSW grouping,
being probably the most homogeneous group of role judges
within the Council (as discussed above under general
79
hypothesis 2), would probably demonstrate more unanimity in
role perceptions than the general category of social work
ers, inasmuch as some social workers (as defined here) have
had no graduate professional training.
Research hypothesis 3. Good role perceivers (those
nearest to the modal pattern) in terms of their own cate
gorical roles tend to be good role perceivers of the rela
ted roles.
The modal pattern, as defined in Chapter I, is a
profile depicting the ma jority-response configuration of
expectations for a categorical role, as set by a group of
role judges, in response to a set of role-cue statements.
A modal pattern was constructed for each categorical role
and each role judge was scored in terms of the modal pat-
tern for the role category in which he has his incumbency.
Good role perceivers are those role judges whose per
ceptual judgments follow most closely the modal pattern of
responses for a given categorical role. (The method by
which this was determined will be discussed later on in
this chapter.) As was pointed out in Chapter II (p. 32),
there is support for the generalization that those persons
^The technique used for the construction of the
modal pattern and the method for scoring the individual
respondent is fully discussed in a subsequent section in
this chapter entitled, "Determining the Modal Pattern."
80
in the group, who approximate most closely the modal pat
tern of role perceptions set by the entire group, are like
ly to have more characteristics in common than those who
deviate greatly from the modal pattern. If this is true,
it would seem justifiable to posit an hypothesis that those
who perceive their own role as others in the group perceive
it are also more apt to view other roles in the manner in
which others perceive them. Moreover, in view of the gen
eralization concerning the closeness of the self role to
that of the other role in role theory, it would seem reason
able to assume that those who perceive the "self" role
clearly can do so only because they have perceived the
"other" role accurately. (See Chapter II, p. 23 ff., sec
tion on "Taking the Role of the Other.")
Research hypothesis 4. (a) Role clarity is greater
among long-term participants than among short-term partici
pants; and (b) role clarity is greater among social workers
than among laymen.
In the questionnaire used in this study, the re
spondent is asked to estimate the degree of clarity with
which he perceives his own role, and also to estimate the
degree of clarity with which he believes the other categor
ical role incumbents perceive their respective roles. Role
clarity, therefore, refers to the respondents perception
of how definite he believes the role expectations are among
81
the incumbents of each of the categorical roles. In the
two parts of this hypothesis, the generalizations are being
made that the variables of "time," of "homogeneity of back
ground," and of "professional training" are important fac
tors in the structuring of role perceptions.
Research hypothesis 5. (a) Role involvement is
greater among long-term participants than among short-term
participants; and (b) role involvement is greater among
social workers than among laymen.
For purposes of this study, role involvement is de
fined as the extent of identification of the Council dele
gate with the Council, as measured by the kinds and amount
of his participation in those activities and concerns of
the Council which are included in the index of role involv-
ment in the questionnaire (discussed at length later). The
rationale for the items utilized in the questionnaire is
this: persons who feel a strong sense of belonging to the
group, and who value that group, act differently from peri
pheral or marginal members of the group. Morever, a strong
sense of belonging denotes a deeper involvement in the role
of incumbency than does marginal attachment to the group.
It is being assumed that these differential actions are
measurable, and that the items in the questionnaire can show
the kinds and amount of Council participation.
Research hypothesis 6. (a) Social workers tend to
82
anchor their role perceptions of the Social Worker and
Staff categorical roles within the professional training
orientation, whereas laymen anchor their perceptions of
the Lay categorical role within the broader societal orien
tation; (b) when perceiving the social worker and staff
categorical roles, those with a Master of Social Work de
gree are more professional-training oriented than are non
master's social workers; and (c) long-term participants
are more intra-council oriented in their role perception
anchorages than are short-term participants.
It was shown in Chapter II that the group to which
an individual belongs, and the group into which he aspires
to be accepted, may provide a frame of reference for his
actions and perceptions. Role anchorages in this tudy,
refer to the groups which the individual uses as a frame
of reference for his views of "self" and his "taking of the
role of the other."
The generalizations contained within the different
parts of this hypothesis would appear to be logical, in that
each is based upon what would seem to be the most likely
source of reference for each of the categories of role
judges. For example, social workers are part of a profes
sion (the Staff member is a social worker by training and
identification also), and it has been shown that the pro
fession to which one belongs, and upon which one's general
ized status is so dependent, probably induce the use of
83
professionally accepted norms and points of view as a point
of reference for group activities, particularly when those
activities are related to that profession. A profession
in the field of social work would seem to serve as a par
ticularly important anchorage point, inasmuch as that pro
fession is a relatively young one. Persons who are fight
ing for a place in the professional "sun" would be expected
to be keenly aware of the importance of acting as a social
worker should act, and by so doing would tend to keep their
behaviors close to the core of expectations for the profes
sion. ^
The layman, on the other hand, having no common
backgroundof training and experience, would seem to have to
draw upon the broader societal norms, and upon his member
ship groups outside of the Council as potential sources for
referents.
It is further generalized, in this hypothesis, that
after having been in the Council for a relatively long per
iod of time, one would come to look "within the Council"
itself for perceptual anchorages. This would include such
factors as precedence and Council sub-groups. Certain
entangling alliances tend to be formed over a period of
years and obligations incurred by favors done to one in the
■^The non-professionally trained social workers
often appear to side with the laymen when certain types of
issues are discussed.
past are not matters to be treated lightly. Such factors
as these tend to crystalize with time, and are apt to be
reflected in the informal structure of the group. Role
theory says that if they are within the structural proper
ties of the group, then they may influence both role enact
ment and role perception.
Research hypothesis 7. (a) The greater the degree
of role involvement, the greater the degree of role clar
ity; (b) the greater the degree of role involvement, the
greater the degree of role consensus; (c) the greater the
degree of role involvement, the greater the degree of
stated Council role satisfaction; and (d) the greater the
degree of role involvement, the greater the degree of
stated Council role interest.
Two concepts have been introduced in this research
hypothesis, namely, "role satisfaction" and "role interest.
Role satisfaction is defined, in this study, as the re
spondents perception of his satisfaction derived from
Council participation, as compared with his most satisfying
community activity. Similarly, role interest is defined as
the respondent's perception of his interest in Council ac
tivities, as compared with his most interesting community
activity. Rather than to permit each respondent to fix his
own referent or anchorage point, he was asked, in this case
to compare his Council role, in both interest and satisfac
85
tion, with that civic activity (in which he had already
engaged) which had proven to be the most interesting of all
his activities, and with the one which had given him the
most personal satisfaction. In most cases, in this study
design, the respondent is asked to select from a number of
potential anchorages the one he deems to be the most im
portant in his own case. Here, however, the matter is more
structured, though not completely so, in that the question
naire specifies the criterion to be used in the selection
proce ss.
All of the segments of this hypothesis deal with the
relationships among variables which, for the most part,
have been used in the other research hypotheses. With re
gard to the first of these (the greater the degree of role
involvement the greater the degree of role clarity), it
would seem reasonable to assume that as one becomes in
creasingly identified with a group, he would come to per
ceive his own role, and the roles of the others, in a more
distinct and clearly defined manner. This might be ex
pressed, for example, in his willingness to spend time in
helping to carry out the functions of the group. Such ex
periences should add up to a greater sharing of views and
judgments with others in the group.
Perhaps the underlying logic of some of the general
izations made in this cluster of hypotheses can be made
more explicit by using one of the criteria included in the
86
role involvement index, specifically the kind and amount of
concern which the respondent has regarding actions taken by
the Council. If, for example, a given action upsets him,
if he talks about it outside of the Council, if he joins
with others to try to do something about it, etc., would it
not seem reasonable to assume that he is more deeply in
volved in his Council role than if he does not give the
action a "second thought?" Likewise, would it not seem
reasonable to assume that if he shows this kind of concern
about Council actions, he is more likely to be interested
in his Council role?
Research hypothesis 8. The greater the degree of
role consensus, the greater the degree of stated role clar
ity.
This hypothesis states, essentially, that role judg
es who express a high degree of unanimity of role percep
tions, will also view the various categorical roles as
being more clearly structured. If the respondent perceives
a given role as others in his categorical role tend to per
ceive it (consensus), the role would appear to be well
structured. Further, if the role is well structured, then
it would seem that those who have learned what the structure
is would perceive it as being clear. The questionnaire asks
specifically for the respondent's estimate of the degree of
clarity with which the role is understood.
Construction of the Questionnaire
87
The role consensus battery
In Chapter II, the concept of "role-cue" as a means
for determining role consensus was presented, together with
a rationale for its use. For purposes of this study, this
concept has been defined as being a normative statement
submitted to the Council delegates for the purpose of de
termining whether a group of role judges would attach it
consistently to any categorical role.
In the questionnaire, role-cue statements were con
structed around three major foci. The first of these con
cerned Council functions and policies (Instrument B in the
questionnaire); the second revealed responsibility for in
itiating action (Instrument E); and the third utilized a
set of simulated quotations with no particular content ori
entation (instrument D), which required each respondent to
attribute each of these contrived quotations to one of the
three categorical role incumbents. As each of these three
instruments required the respondent to make a different
kind of response, each will be discussed separately. The
reason for using three different kinds of approaches was to
make it possible to use the results of the three instru
ments as reciprocal checks upon one another, and thus to
increase the reliability of the total findings on role con
census among council delegates. These three instruments
88
constitute the core of the questionnaire, and are referred
to hereafter as the "Role Consensus Battery," which will
be described in detail.
Although three different approaches are used to
measure consensus in this study, continuity is maintained
throughout the Battery. For example, each respondent was
asked to indicate his perception (for each item in each
instrument) of the role which he himself plays (role of in
cumbency), and also to indicate his estimate of how the
other role incumbents perceive their own roles. Addition
ally, a pattern of similar approach was utilized throughout
the Battery, inasmuch as preceding the directions for fill
ing out each of the respective instruments, an introductory
statement was made concerning the intent of that instru
ment. It was assumed that such a statement might help to
standardize, to some extent, the reference range used by
the respondent to the questionnaire item.^ However, there
are several basic differences in the three instruments
which will be evident from the following analysis.
Instrument B. This instrument, built around Council
functions and policies, is set up on an "agree-disagree"
basis (see Appendix A, p. 210). The respondent is asked to
indicate whether he perceives each one of a group of state-
-^Something of this same assumption influenced the
composition of the cover letter which accompanied the ques
tionnaire. See Appendix A, p. 225-26.
89
merits as being one with which those playing the role would
be likely to agree or disagree. He is also asked to do this
for each of the two other categorical roles represented in
the Council. In addition to this, and in an effort to help
him grasp the concept of "role" as such, he is asked to in
dicate his personal reaction to each statement. This ena
bles us to see if there is any relationship between incum-
bency-role perception and a personalized perception of a
self role. Instrument B, therefore, asks each respondent
to make four role perception judgments on each of the four
teen items which comprise the instrument.
One other facet has been built into Instrument B
differentiating it further from the other two instruments
in the Battery. This will be treated in detail in the sec
tion entitled "Modal Pattern Responses— Axial Orientations,"
and hence needs only be mentioned at this time.
Instrument D ("Who Said That?").— This instrument
asks the Council delegate to attach a simulated quotation
to one of the categorical roles (see Appendix A, p. 213).
"Of whom, in your judgment, would they be most typical or
characteristic?" is included in the prefatory statement of
the instrument. And again, in the directions for the in
strument, is included this statement, "... which one
(role occupant) would be most likely to have made such a
statement?" This instrument follows most literally the
90
concept of role-cues as used by Sarbin and W i l l i a m s .
Forced choice is avoided as the respondent may, if he
chooses, check a column headed "none" for his choice.
Instrument D contains thirteen items. One statement was
included which is based on objective fact, namely, a con
stitutional provision. This was included as a possible
check of the respondent*s factual knowledge about the Coun
cil's formal structure (item # 10).
Instrument E (Role responsibility).— This set of
items specifies seventeen possible functions or actions,
which presumably could be or are carried on in the Council,
and asks the respondent to assign primary responsibility
for initiatory action in terms of the various categorical
roles. The respondent is not only asked to assign "most
responsibility" but also "least responsibility" (see Appen
dix A, p. 214). In the analysis of the data, the responses
can consequently be treated either in dichotomous fashion
("most" and "least") or in a ranked, one, two, three, man
ner, inasmuch as there are three categorical roles involved.
The three instruments, discussed above, calling for
a total of 103 responses, constitute the tools developed
for role consensus measurement, and together are to be re
ferred to as the Role Consensus Battery. The remaining
^-®See Chapter II, pp. 31-34, for a discussion of
this technique as it was introduced by Sarbin and Williams.
91
portions of the questionnaire were constructed to provide
supplementary and supportive information to the data on
role consensus. It remains to be described how the Battery
instruments were used to interpret this subject of role
consensus.
Determining the modal pattern.--It should be pointed
out that in this study little concern is being given to the
delineation of the boundaries of the role functions, or the
range of behavior expectations for the various roles. Con
sensus alone is being treated here; and, further, consensus
only insofar as the analysis of certain role-cues will per
mit. There are no implicit criteria as to what constitutes
good or adequate role performance, or even what the role
incumbents believe to be good role performance. The Coun
cil delegates have simply been asked to respond (1) to
certain role-cue statements alleged to have been made by
Council members, (2) to expressions about certain responsi
bilities, and (3) to some statements concerning Council
functions and policies, and to classify these statements in
terms of their own perception of the various categorical
roles in the Council. These responses permit us to con
struct a profile or modal pattern for each categorical
role. It is upon this basic modal pattern that many of the
analyses rest.
In constructing the modal pattern, it was first
determined which alternative choice for each item in the
Instrument received the majority of actual choices within
the grouping of role judges— MSW, Non-MSW, or Lay. Consen
sus was interpreted as being the majority of the actual
choices, and whether this was, in the case of Instrument B,
for "agree" or "disagree" was immaterial for purposes of
this study. However, the different instruments presented
different kinds of classifications as to what constituted
a majority.
Instrument B, being dichotomous in its structure,
would require that anything over fifty per cent be consid
ered as consensus for each item in that instrument. How
ever, Instruments D and E were trichotomous in structure,
i.e., each questionnaire item could be attached to any one
of three categorical roles. For Instruments D and E,
therefore, any potential alternative choice which received
more than one-third of the actual choices of the role
judges for which the modal pattern was being constructed,
constituted consensus for that instrument item. Conse
quently, the lower limit of consensus for Instruments D and
E became 34 per cent of the actual choices for each item,
whereas for Instrument B it became 51 per cent of the actu
al choices made. This process yielded a modal pattern for
each role judge grouping of council participants, as they
perceived each of the categorical roles (and on one occa
sion, a modal pattern for their personal role).
93
A basic assumption of the next procedural step, and
one which should be recognized at the outset, is that the
extent of the general consensus on a given item should be
taken into account when viewing case deviations from the
modal patterns or congruencies with it. It is further be
ing assumed that deviations or congruencies are more sig
nificant if they occur on items of high consensus than on
items of low consensus. Weighting each item in terms of
its place on a consensus scale makes it possible to take
this fact into consideration when treating the data.
Therefore, the next step in constructing the modal pattern
for each of the three groupings of role judges, was to lo
cate the degree of consensus for the given item on a scale
and then to weight it appropriately.
One hundred per cent, the maximum agreement poten
tial, became the upper limit for the weighting scale. The
lower limit was automatically set— bl per cent or 34 per
cent, depending upon whether the alternative choices were
two or three. Inspection of the data seemed to suggest in
tervals of 10 per cent, between the upper and lower limits,
as being reasonable for weighting purposes, since it would
provide a range that would not be so broad as to loose sig
nificant differences, or so fine as to create artificial
ones. Weights were arbitrarily assigned to each of these
step intervals as follows: for Instrument B, weights of one
to five were used as there are five step intervals (of 10
94
Per cent each) between 51 per cent and 100 per cent; for
Instrument D and Instrument Ef weights of one to six were
used, since there are six such step intervals between 34
per cent and 100 per cent. The following summarizes the
consensus step intervals and the corresponding weights as
signed to each;
For Instrument B
Per Cent Consensus 51-59 60-69 70-79 80-89 90-99
Assigned Weight 1 2 3 4 5
For Instruments D and E^-9
Per cent Consensus 34—43 44-53 54-63 64—73 74-83 84-93
Assigned Weight 1 2 3 4 5 6
The following example will show how this procedure
has been used to arrive at a given MSW respondent's consen
sus score, using statement #7 in Instrument B as an illus
tration. First, as over 90 per cent of the MSW role judges
checked "disagree" for this statement, it is clear that
consensus is disagreement with the statement. Next, the
extent of the disagreement is used as the basis for assign
ing a weighted score to this item. Inasmuch as the disa
greement among the MSW role judges was high (over 90 per
^It should be pointed out that the scores derived
from these two weighting schemes were never made additive
in the analysis of the data.
95
cent) the weighted score of statement #7 is to be found in
the 90-99 per cent category. Therefore, for MSW role judg
es, statement #7 is assigned a weighted score of 5. Thus,
when our sample MSW1s response to statement #7 follows the
modal pattern ("disagree") he receives a score of +5 for
statement #7. If, however, he does not follow the modal
pattern (he checks "agree"), then he receives a score of -5
for statement #7. The respondent's modal score, then, for
Instrument B is determined by summing his weighted scores
on all of the statements in Instrument B.
The above procedure makes two different kinds of
measurement possible: (1) a total comparative consensus
score for each categorical role which constitutes a modal
pattern for each of the three council roles; and (2) a com
parative consensus score for each respondent showing his
approximation to the modal pattern.
Although, as stated previously, the Battery consti
tutes the core unit of the questionnaire, there are a num
ber of sections which remain to be described. They will be
dealt with subsequently.
Questionnaire sections with intra-Council focus
Three sets of items in the questionnaire have been
given an intra-Council directionality (i.e., they point to
ward factors within the Council): (1) estimated clarity of
categorical role perceptions, (2) personal satisfaction
96
from role playing, and (3) extent of role involvement. An
explanation of each of these sets of items follows.
Clarity of role expectations.--In section "C" of the
questionnaire, a straightforward question was asked about
the perceived clarity of Council role expectations (see
Appendix A, p. 212). In general, the same pattern has been
employed here as in the Role Consensus Battery. One addi
tional facet, peculiar only to Instrument B, has been re
tained. Both social workers and laymen were asked to esti
mate, on a four-point scale (with no intent to imply that
the points necessarily represent equidistant locations),
the clarity with which they perceive their own roles, and
also the extent to which they think other role incumbents
perceive with clarity their respective roles. Here, as in
Instrument B, it was thought to be helpful to the respond
ent to differentiate between "role patterns" and "individu
al responses," as such, by also asking him to estimate the
clarity with which he perceives his own personal role.
This was an attempt to learn whether a role incumbent dif
ferentiates between himself as a role player and the other
persons in his own role category, as they play their roles.
For example, would one find any support for the notion that
a given social worker might perceive his own role as being
clearly delimited, but perceive other social workers as
being confused as to their own role perceptions? The
97
scale used in this section sets the outer limits from
"usually clear" to "usually unclear," with two intervening
points, "more times clear than unclear," and "more times un
clear than clear."
Role involvement . — In this study, it has been hy
pothesized that both clarity of role perception and role
consensus would be related to the degree to which Council
delegates are involved in their respective Council roles.
Involvement in the role is defined as the extent of identi
fication of the Council delegate with the Council, as
measured by the kinds and amount of participation in the
activities of the Council as reflected in the participation
(involvement) index.
It was reasoned that if an individual spends a con
siderable amount of time in the work of the Council, or if
some of the decisions made in it disturb him enough to
cause follow-up action, or if the Council is referred to
with the use of the personal pronoun (we, our, us, as
against they, them, it), then he is likely to be more deep
ly involved in his role than if the reverse of the above
variables is true.
In the construction of the index, no assumption is
made that the items included represent a scale, i.e., that
the items represent any ascending or descending order.
McCormick and Francis, in their discussion of scales and
98
and indexes, make the following distinction:
It should be noticed that an index is something
quite different from a scale, although it also assumes
some common denominator. The common denominator, how
ever, is by no means the same thing as a single dimen
sion. 20
All of the items included in the index definitely refer to
some form of activity that is Council-related; no item was
included simply because it might correlate with Council
participation. Consequently, the possibility that contra
dictory elements were included in the index does not appear
to be very great. Further, it does not seem likely that
the problem of scalability is involved.
The following is a summary of the kinds of Council
participation included in the index (see Appendix A, p.
221): attendance at regular Council meetings; serving on
committees or performing special assignments, including
whether these were the result of volunteering for the tasks
or of being asked to serve; statements as to whether Coun
cil work has been undertaken at the expense of other valued
activities; quality of participation in the Council delib
erations; the extent to which Council decisions or actions
disturbed the respondent, and the manner in which he fol-
lowed-up (or failed to follow-up) his disturbance; state
ments as to whether Council matters enter into casual
20Thomas C. McCormick and Roy G. Francis, Methods of
Research in the Behavioral Sciences (New York: Harper &
fer othe r s, 19b8 }, p'. iu7.
99
conversations with persons outside the Council; the extent
of identification with Council actions as this is reflected
in the use of pronouns such as "we" versus "they" when re
ferring to such actions; and the defense of Council actions
to persons or groups outside of the Council.
After selecting the above kinds of actions which
showed promise of indicating belongingness or identifica
tion with the Council, the next step was to assign weights
to the respective items.
Although Haygood and Price indicate, "Whenever some
objective technique of weighting items can be utilized, the
results are generally to be preferred to those obtained
pi
with weightings assigned from subjective judgments,"
certain conditions prevailed in this study which seemed to
make it impossible to avoid the arbitrary assignment of
weights to the index items. One of the recognized objec
tive techniques for assigning weights is the use of inde
pendent judges. This did not appear feasible in this
study, as the only adequate judges were the members of the
Council themselves, and it did not seem reasonable to have
the persons being tested serve as the judges of the item
weights. Further, as was indicated in the section of the
pretest of the questionnaire, the group used for pretest
“ ^Margaret J. Hagood and Daniel O. Price, Statistics
for Sociologists (New York: Henry Holt and Company, 19b2J,
p - . r ~ ~
100
purposes was not sufficiently like the measured group to be
likely to attach the same meanings to the different index
items as the actual Council situation would warrant. Con
sequently, it seemed necessary to assign the weights arbi
trarily and to rely on certain checks, subsequently dis
cussed, to justify the assingment. An abridged statement
of the items in the Role Involvement Index follows, togeth
er with the weighted score (in parenthesis) given to each
00
index item.
Item
1. Attendance at regular meetings: Most, (b); Half,
(3); Seldom, (1).
2. Served on standing committees: One, (1); Two, (3);
More than two, (b).
3. a. If served on standing committees, total number
of meetings attended: Less than three, (1);
Three to five, (2); More than five, (3).
b. If volunteered to serve, (b).
c. Served on any special assignments; time devo
ted to such equivalent to: Three committee
meetings, (1); Three to five, (2); More than
five, (3).
(All items under #3 were treated as a unit. The
final score given to the respondent was 1/3 of the
total for item #3-a, b, and c. This was done to
prevent giving too much emphasis to this phase of
council work and yet to take into account some dif
ferent aspects of such work.)
4. Has your Council work necessitated your dropping
22See Appendix A, Instrument G, for a complete
statement of each index item as it appeared in the ques
tionnaire.
101
any other activities or kept you from taking on
some activity which you would have otherwise
taken on?: Yes, (5); No, (0).
5. Some kinds of participation in Council meetings:
Frequently involved in discussion or debate, (4);
Occasionally, (3); Ask an occasional question,
(2); Listen and vote, (1).
6. Feelings about council participation: Often feel
that my experience and position justifies speak
ing up, (4); Occasionally feel that my position
is relevant to issues at hand, (3); Often feel
that I should speak up but hesitate, (2); Seldom
feel I have anything to contribute, (l)j Usually
at a loss to know what is going on, (0).
7. Have there been council decisions or actions which
disturbed you or given you cause to be really
"bothered?": Yes, (2), No, (0).
If answer is "yes," which best characterized
your subsequent actions?
a. Followed up, sought more information, (3).
b. Got in touch with someone who could do
something, (3).
c. Joined with others to try to do something
about it, (3).
d. Brought the matter up at a subsequent Coun
cil meeting, (2).
e. Decided that time would take care of the
matter, (1).
f. Upset at first, but dismissed it subse
quently, (0).
8. Are matters dealt with in the Council ever brought
up for discussion by you in information conversa
tions with your friends?: Frequently, (3); Occa
sionally, (2); Seldom, (1).
A. If answer is "frequently" or "occasionally,"
which best characterized your reason for
doing so?
a. Think it part of my job to let them
know what is taking place in the
council, (3).
b. My friends are interested in what the
council does, (3).
102
c. Friends can help me think through some
matters on which I am called on to
vote, (3).
d. Such matters make good conversation,
(0) .
e. Like my friends to know that I am in
volved in serious community mat
ters, (0).
B. If answer to #8 is "seldom" or "never,"
which best characterized your reasons for
so doing?
a. Council matters should stay within the
council, (0).
b. Friends would not be interested, (0).
c. Friends would not be impressed with
council matters, (0).
d. When I leave the council, I shut coun
cil matters out of my thinking, (0).
9. When referring to some action of the council of
which you approve, which set of pronouns would
you prefer touse?
Set "a"— "We, our, us," (2).
Set "b"— "They, their, it," (0).
When referring to some action of which you dis
approve, which set of pronouns would you prefer
to use? Set "a", (2); Set "b", (0).
10. Do you ever find yourself in a position of volun
tarily defending the council or its actions
outside of the council? Frequently, (3); Occas
ionally, (2); Seldom (1); Never, (0).
If your answer is "frequently" or "occasionally,"
among which of the following persons or kinds of
groups is this most likely to happen?
(Nine different categories are listed from
which to choose. Two points were given for
checking any one category; one point was added
for each category checked thereafter.)
If your answer to #10 is "seldom" or "never,"
among which of the categories listed would you
mostly refrain from defending the council or its
actions? (No points were given for this item.)
103
It is recognized that this method of arbitrary as
signment of weights to index items has certain limitations.
Obviously, a given score might be possible with many dif
ferent combinations of index items. Further, one would
have no right to assume, for example, that the difference
between a score of 20 and 25 would be equal to the differ
ence between a score of 10 and 15. In compiling the data
from this study, care was taken to avoid making such assump
tions. On the other hand, it was believed that the homo
geneity of the items selected for the index (discussed
above), together with the conscious effort that was made to
keep any one item from overshadowing another by overweight
ing it, would result in the index providing a reasonably
accurate means for determining role involvement, as defined
for this study.
Some assurance of the reliability of the above de
scribed weighting system is to be had from the results of
the following test which was applied to the index. Jahoda,
Deutsch, and Cook suggest that when it is not feasible to
use the test-retest method, or two equivalent measures, it
is possible to split the test which has already been given
into two equivalent halves, and determine if there is any
correlation between the scores on the two halves. This
^Marie Jahoda, Morton Deutsch, and Stuart W. Cook,
Research Methods in Social Relations (New York: The Dryden
•Press ,""19T ol),, P ~ € “ l, p. 1U4-1U5.----
104
is called the split-half method. These writers, in de
scribing the test, indicate its use as follows:
Comparison of each individual's scores on the two
halves of the test is essentially a procedure for de
termining whether the items in the test all tap the
same attitude or characteristic— i.e., whether they
represent the same universe of content.24
They further indicate that the two halves of the test may
be determined by using all the even-numbered items in one
half and the odd-numbered items in the other half, "...
although any method of dividing the test or measurement
into equivalent halves is u s a b l e . "25 Following this sug
gestion, the role involvement items (section G in the ques
tionnaire ) were divided into equivalent halves by selecting
alternate items in the questionnaire for each of the two
halves. The resultant correlation of the scores made by
all role judges on the two halves was then determined to be
.523 (r » .523). Guilford suggests that:
The full-length test is not twice as reliable as
the half test, but its reliability is greater and can
be estimated by the special Spearman-Brown formula
with n * 2.26
Walker and Lev also suggest the use of the Spearman-
Brown formula, as a corrective for determining "the corre
lation which would have been obtained between scores on two
Ibid., p. 105. 25ibid., p. 104.
26j. p. Guilford, Fundamental Statistics in Psychol-
ogy and Education (New YorkT McGraw-hiill Bo^ok Company, Inc. ,
1956), p. 452.
105
full test forms had such been available. "27 The formula
28
for the Spearman-Brown method, as indicated by Guilford,
is:
2rhh
-hh
1 4 rhh
He also indicates that an estimate by the use of the
Spearman-Brown formula is probably conservative, as it
tends to be an understatement.
The use of the Spearman-Brown method yielded the
result of r ^ a .69. This figure provides a measure of the
reliability of the total role involvement index as estim
ated from the .reliability of the halves to each other. How
ever, the figure does not tell how much variation in the
two halves is accounted for by their mutual relationship.
The latter can by shown by Squaring the result ob
tained from the Spearman-Brown formula (.69) yields .48.
From this figure it can be conservatively assumed that 48
per cent of the total variation in the two halves of the
role involvement index is explained by their common varia
tion or mutual relationship. To state the matter differ-
2^Helen M. Walker and Joseph Lev, Statistical In
ference (New York: Henry Holt and Company, l^bs), p. 3T0.
2®Guilford, _ ojd. cit. , p. 452.
2^Morris Zelditch, A Basic Course in Sociological
Statistics (New York: Henry Molt and (Company, l95£j, p. 108
106
ently, it would seem conservative to assume that 48 per
cent of the variability in role involvement of Council role
judges can be accounted for by items in the role involve
ment index. This seems to indicate that the index does
have homogeneity and internal consistency, and that there
is justification for its use in this study.
Role satisfaction and interest.— Satisfactions which
come from role playing are, in the main, outside the pur
view of this study. However, it was believed that touching
upon them briefly might throw some light upon role consen
sus and role clarity (see questionnaire, Appendix A, Sec
tion F-2, p. 217).
Both interest and satisfaction are relative terms,
and for this reason respondents were asked to use a common
base in judging their interest in, and satisfaction from,
Council role playing, namely, to compare their Council role
with the civic activity (in which they had already engaged)
which had proven to be the most interesting of their activi
ties, and the one which had given them the most personal
satisfaction. This comparison was stated in terms of four
specified alternatives: "more than the best," "about the
same as the best," "somewhat less than the best," or "so
much less than the best that there is no comparison."
Questionnaire sections with
extra-Council focus
107
Do laymen and social workers tend, in any patterned
way, to anchor their role perceptions in out-of-Council
roles of functions which may, in turn, throw light on
Council role consensus? Three sections of the question
naire deal with (1) out-of-Council influences on Council
role behavior; (2) reasons for willingness to spend time in
Council work; and (3) the extent to which the Council par
ticipant would like his Council participation known in
circles outside the Council.
Recognized role influences.— In section F-3 of the
questionnaire (see Appendix A, p. 218), the respondent was
asked to select three potential out-of-Council role influ
ences and to place them in rank order of their importance.
These possible influences, from which his choices were to
be made, start with the broader societal norms, then move
to the professional-education and the professional-practice
norms. The pattern adopted in the Role Consensus Battery
is repeated here; each respondent chooses for himself (per
sonally), chooses for others who share the same categorical
role as himself, and chooses for the related roles, accor
ding to his perceptions of how those in such roles would be
likely to choose. In all, then, the respondent here makes
four sets of rank order choices (first, second, and third),
one set for himself, and one set each for the Social Worker,
Layman, and Staff roles— a total of twelve choices for this
108
section.
Reasons for doing Council work.--Closely related to
the above, but coming at the matter somewhat differently,
the attempt is again made to discover the delegate's link
age of his Council role with his out-of-Council roles,
positions, and self-estimate criteria (section F-l, Appen
dix A, p. 216). The respondent is asked to show how much
consideration in evaluating his Council role (major, some,
or none ) he would ascribe to certain selected factors which
cluster around two different axes: (1) criteria for self
judgement, such as "contact with important people," "per
sonal satisfaction," "usefulness as a citizen," and "use of
leisure time;" and (2) expectations anchored in out-of-
Council roles, such as "expected of one in my kind of work,"
"common practice among my associates," and "expected of the
spouse of a person who wants to be prominent in the commun
ity. "
Desire to have participation known.— In the matter of
potential out-of-Council role linkages or anchorages, a
further assumption is made that if the respondent perceives
his Council role as being potentially useful to him in per
forming some of his out-of-Council functions, then he will
want his participation in the Council to be known to per
sons related to his out-of-Council roles. In section F-6,
(see Appendix A, p. 220), ten kinds of associations are
109
listed, and the respondent is asked to react to each in
terms of the following: "would like them to know" (i.e.,
about my Council participation), "prefer they not know,"
"makes no difference" (whether they know or not).
Pretest of the Questionnaire
The unique structural elements of the Council has
made it impossible to find an identical group locally for
pretest purposes. Structurally speaking, the other coor
dinating councils (Leisure Time and Health) are identical,
but the content of their operations is obviously different;
consequently, the questions which apply to the Council are
not relevant to them. The best available group, for this
purpose, was one where the content was similar, and the
structural properties sufficiently similar, to justify its
use.
The pretest group chosen was the Executive Committee
of the Family and Child Welfare Division (Minneapolis,
Minnesota), a section of the Community Chest and Councils
of Hennepin County. This Executive Committee, although much
smaller (19) than the St. Paul group, is similar in fre
quency of meetings and work content. The major differences
are that the Minneapolis group spreads the base of its de
cision-making much less broadly than the Council claims to
do and that its membership, although composed of both lay
men and social workers, as is the CouncilTs, contains a
110
greater percentage of persons from the upper status posi
tions in the welfare community, i. e.f agency executives
and prestige laymen.
The data obtained from the pretest group, therefore,
had only nominal value as a check upon the results of the
study proper. However, it did perform one important func
tion, namely, that of making possible a check on the com
position of the item statements themselves and the direc
tions given to the respondent. It revealed which state
ments were not adequately comprehended, and provided a
check on the format of the questionnaire in actual use.
Permission was given by the Executive Director of
the Community Chest and Councils of Hennepin County to use
the Executive Committee as a testing ground for the ques
tionnaire to be used in this study. The writer worked
through the Director of the Executive Committee who was
most helpful. Each member of the Committee was contacted
by telephone before being sent the trial questionnaire. The
nature of the main study was explained to him, as was the
function of the pretest procedure. The respondent was then
asked to make whatever written comments, suggestions, or
responses he deemed appropriate to any and all aspects of
the questionnaire. Fourteen well annotated responses were
received from the nineteen members of the Executive Commit
tee. This pretesting was carried on during September,
1958.
Ill
The responses of the pretest aided considerably in
establishing confidence in the general format and approach
of the questionnaire, although it also brought about some
substantial modifications in the questionnaire. These re
lated to size reduction, realignment of some items, and
the recomposition of some items and some parts of the
directions to the prospective respondent.
Statistical Tools Utilized
The use of nonparametric
techniques for data anaTysis
In planning the design for the study, it was appar
ent at the outset that some of the more conventional stat
istical tools might not be appropriate to the data results
being sought. It was decided, instead, that nonparametric
statistical techniques gave promise, in the main, of being
more suitable to the analyses called for in this study for
four main reasons.
(1) There did not appear to be sufficient grounds
for making statistical inferences requiring the assumptions
about the conditions of the parameters of the population
which are inherent in the parametric statistics. Siegel
points out that nonparametric tests need make no assumptions
about shape of the population.30 However, nonparametric
30sidney Siegel, Nonparametric Statistics for the
Behavioral Sciences (New York: "McGraw-Hill Book^Company7
Inc. , 1.956 ), pp . 1-5 and 30-32.
112
tests axe not limited to such situations. They can also be
used when the shape of the population can be safely as
sumed. While it is true that they axe xated as less power-
ful than paxametxic tests, the factox of powex-efficiency
of paxametxic statistics is neutralized if the shape of the
population cannot be assumed.3- 1 - Thexefoxe, it appeaxed
that the use of nonpaxametxic tests in.this study pxovided
gxeatex pxobabiiity of producing dependable data, even in
the light of their power limitations, than would the use of
paxametxic tests which specify certain conditions about the
parameters of the population.
(2) When working with a relatively small N, such
as was the case in this study, it seemed particularly nec
essary to avoid statistical inferences where the population
distribution could not be assumed.
(3) The nature of much of the data produced by this
study, suggested the use of ordering or ranking approaches
rather than measurement per se. A statement from Siegel
may help to sharpen this idea.
Nonparametric statistical tests are available to
treat data which are inherently in ranks as well as
data whose seemingly numerical scores have the strength
of ranks. That is, the researcher may only be able to
say of his subjects that one has more or less of the
31»Power-efficiency" of a test refers to the amount
of increase in sample size which is necessary in order to
make that test as statistically discriminating as is the
most efficient test known of its type. See Siegel, ibid.,
p. 20.
113
characteristic than another, without being able to say
how much more or l e s s .32
In this study, for example, role judges were asked
to attach certain role-cue statements to one of three
categorical roles. There is no basis for saying "how much
more" a given individual attaches a given cue to a given
role than does some other individual. The only basis for
saying "how much more" is restricted to the number of per
sons who accept or reject the cue as belonging to a given
role. The Sign Test, discussed later, which was used in
this study, is a case in point. It calls simply for the
use of plus and minus values rather than for quantitative
measures as such. It was believed that this type of test
served the purposes of this study more reliably than a test
which required true numerical values in the usual statisti
cal sense.
(4) Another advantage of the nonparametric method
is the relative ease with which it can be administered.
It must be recognized, however, that nonparametric
techniques are not without limitations, the principal one
being power efficiency, which was mentioned and defined
above. Reduction in power necessitates increasing the size
of N in order to yield a high order of reliability. From
a practical point of view, this means that when a probabil-
^2Siegel, ibid., p. 33.
114
ity table Is employed where the N is small, the evidence
must be almost overwhelming (as is the case with the Sign
Test), before it becomes significant at the .05 level.
Obviously, on the other hand, if the data permit the assump
tion of a normal distribution of the parameter, statistical
significances can often be claimed without being so statis
tically demanding. In other words, it is possible that
nonparametric techniques may demand the discarding of data
as not being significant when, in reality, they might be
important or even significant given the assumptions of a
normal distribution of the population.
Statistical techniques used
in the studyr
The raw data resulting from the tabulations of the
questionnaire responses were first examined percentage-wise
in terms of the various research hypotheses. These per
centage distributions were then used in three ways; (1)
as a basis of comparison where the categories of respond
ents were of unequal size; (2) as a basis for revealing
trends of the data, which were subsequently tested for sig
nificance by using one of the statistical techniques
adopted for the study (discussed later); and (3) as the
basis for the construction of the modal patterns.
The Sign Test.— One of the statistical tools within
the nonparametric framework is the Sign Test. This is
sugges-ted by Sie^c * fa:
see ms u nvva rranti.ci, aui
"the da ta and to ..aa c
Siegel further indic-it
and Lev,^4 that:
The only as:
variable under
t ion - The r v v ■
form of the d^
assume ^hat a*
lation.
A l t h o u g h t h - .
mo st p o v ' j e r i u . . - - ,
well suited to ci^tj
j udg rne rit be t vve e : i a
The Sign Tv
that the medlar* of
is zero. The n u j . . .
* * i : •
(X. >
' S * .
Klilssfffitf
■ * » j , + ^ , ?
V - r
,X. <
The Sign Test may
test. This stud v
33
34Helen ,i. '.hi
e nee (New Y or k : Non r /
J^Siege1 ,
36.
37
Walker arid
Siege I , or
1 > ‘ ' n
A f 1 j * .
i(W\ P ! >
, > ' i
HP '■ !‘H '
T>!
%
115
suggested by Siegel fox data where quantitative measurement
seems unwarranted, but where it is still possible to rank
o o
the data and to make comparisons of two such rank orders.
Siegel further indicates, and this is supported by Walker
and Lev,^4 that:
The only assumption underlying this test is that the
variable under consideration has a continuous distribu
tion. The test does not make any assumptions about the
form of the distribution or differences, nor does it
assume that all subjects are drawn from the same popu
lation.^
Although the Sign Test is not considered to be the
most powerful tool, Walker and Lev feel that it is, " . . .
well suited to data for which measurement is difficult but
judgment between a pair of objects is possible.
The Sign Test is designed to test the hypothesis
that the median of the differences between two conditions
is zero. The null hypothesis is
p (XA> XB) = p(XA < XB) =-!S.37
The Sign Test may be used either as a one- or two-tailed
test. This study employed it as a one-tailed test, as it
33lbid.t p. 68.
34Helen W. Walker and Joseph Lev, Statistical Infer
ence (New York: Henry Holt and Company, 1^S>3 J, p. 4^6.
^Sie gel, jD£. cit. , p. 68.
^Walker and Lev, jojo. cit. , p. 431.
^Siegel, t cjo. cit. , p. 68.
116
was not only hypothesized that there would be certain dif
ferences, but the directionality of these differences was
also indicated in the research hypothesis, e.g., 3+ > 2-.
The following statements constituted the process-
steps by which the Sign Test results were computed:
1. Numerical distributions for the "agree" and "dis
agree" responses for each categorical role were determined.
2. The majority response constituted consensus.
3. The numerical distributions were translated into
percentages to make possible comparisons of unequal N’s
among role judge groupings.
4. In comparing the various categorical roles by
Council tenure (3*- and 2-), a plus sign value was assigned
to a questionnaire item wherever the percentage supported
the directionality stated in the research hypothesis, e.g.,
3*. y 2-. Thus, if the percentage pointed in the direction
contrary to the hypothesis, the item was assigned a minus
value, or in the case of a tie percentage, the item received
a zero value.
5. Summation of the plus-minus values for all items
in the instrument, for each categorical role being per
ceived, followed.
6. Probabilities associated with such a plus-minus
distribution were determined by utilizing a table designed
for this p u r p o s e . ^3
^^Siegel, ibid., Appendix, Table D, p. 250.
117
Spearman Rank Correlation Coef f icient.— As it was
possible to rank the individuals in two ordered series, in
terms of their responses to many of the variables under
consideration, the Spearman Rank Correlation Coefficient
(Rho) was chosen as being a measure of relationship which
was apropos to the data. Siegel accepts the Spearman Rho
as a nonparametric measure of correlation, and states its
power efficiency to be about 91 per cent of the Pearson r.^9
One of the problems likely to arise when using this
technique, however, is that of dealing with tied ranks. If
the proportion of tied scores is large, the effect is to
increase the value of Rho, and thus bias the results of the
test. The correction factor, which Siegel provides to com
pensate for this, was applied where the highest proporation
of tied ranks was found to occur. (See Table 4, page 139.)
Here the uncorrected Rho critical value was .296 while the
corrected value was .290. This difference between the cor
rected and uncorrected Rho critical value was so slight
that it did not appear necessary to continue to apply the
technique to the remainder of the data.40
The Spearman Rho technique was applied to the hypoth
eses dealing with consensus, where the respondent’s modal
O^siegel, ibid., p. 213.
40see Siegel, ibid., p. 284, for support of this
interpretation of the difference.
118
score was used as the base. (See pages 91 ff. for details
on Modal Scores.) Respondents were ranked on the basis of
modal scores and the standard procedure for computing the
Spearman Rho was then followed.4- ' -
The Chi Square Test.— Although usually associated
with parametric statistics, competent authorities in the
field of statistics adjudge the Chi Square test as appro
priate to data, even though the shape of the population may
not follow the normal distribution model. Walker and Lev,
in introducing the subject of nonparametric methods, state
categorically the appropriateness of the Chi Square test
as a nonparametric instrument.
Most of the methods presented . . . depend upon the
assumption that samples have been drawn from a normal
population. In many problems, this assumption looks
quite unreasonable. . . . Several methods will be de
scribed for making inferences without any assumption as
to the form of the distribution in the population.
Such statistical methods are called non-parametric or
distribution free. Examples of non-parametric methods
. . . are the x^ test, percentiles, and the rank-corre-
lation coefficient.42
Adoption of the Chi Square test was made only after
considerable exploration of the potentialities of other
nonparametric techniques. It was selected, however, as be
ing most appropriate to the analysis, particularly in view
of Walker and Lev's statement that, "The data need not even
41Ibid., pp. 204-206.
42yyalker and Lev, j3£. cit. , p. 426.
119
be of the sort which permit ranking, but may be strictly
categorical. 1,43
Occasionally it became necessary to combine the data
of two related cells, in order to conform to the demands of
the Chi Square test that, when there is more than 1 df, not
more than 20 per cent of the cells can have less than five
cases.44 Moreover, in a few instances, the necessity to
limit the number of small cells required the use of the
2 x 2 contingency table, authority for which is to be
found in Guilford.4^ He recommends, however, that the
Yates's Correction for Continuity be used with the 2 x 2
table in those cases where any expected cell frequency is
less than ten.4^ The application of this technique was
necessary in only two instances, as in ail other cases the
combining of ceils provided ceil frequencies which were
greater than ten.
Percentage distributions.— There were a number of
instances in the study where the use of the more refined
statistical tools, discussed above, revealed no statisti-
43Ibid., p. 447.
44Siegel, oj£. cit. , p. 178.
45
J. P. Guilford, Fundamental Statistics in Psychol
ogy and Education (New York: A/icGraw-Hi 11 Book^tompany,
Inc., i(bbj, p7 236.
4^Guilford, ibid., p. 234.
120
cally significant trend (at the .05 level), but where the
original percentage figures provided possible important
cues for future research. Since research has an obligation
to reveal the total findings so that those who seek to
buiid upon it may be spared the necessity of repeating more
exploratory groundwork, the percentage distributions, where
they suggest possible "take off" points for the formulation
of hypotheses for future studies, have been included.4^
This was particularly true wherever consistency between the
findings of the different instruments seemed to be sug
gested.
The significance level.--As stated in Chapter I, the
significance level was set at .05 even before the data were
collected. The function of the level of statistical signi
ficance is to determine the likelihood of committing a Type
I error (rejecting a null hypothesis which should not be
rejected), so it was necessary to set an appropriate cut
off point where the findings would be accepted or rejected.
In other words, this question must be answered: At what
level is the data to be considered statistically signifi
cant? Statisticians apparently agree that this decision
must be based largely upon the use to which the data are to
47por support of this position, see Robert K. Merton,
George G. Reader, and Patricia L. Kendall, The Student Phy
sician: Introductory Studies in the Sociology of1 Medical
Education l Cambridge: Harvard University Tress, xybVj,"
pp. ^Os-304.
121
be put. Guilford discusses the necessity for being sure,
and at the same time for not setting the significance level
so high that few differences and relationships are accepted
as established. He concludes:
Some kind of balance must be reached. Considera
tions external to the data themselves should be given
weight. There may be serious theoretical or practical
reasons why it would be costly to make one kind of er
ror or the other. Thus, the odds, ultimately, cannot
be decided on statistical grounds.
Inasmuch as important policy decisions were not
likely to be based upon the findings of the study, and also
in view of the exploratory nature of this study, it did not
seem necessary to set as high a level as .01. On the other
hand, a level no more demanding than .10 appeared too low
to provide a base for drawing conclusions. The .05 level
was selected as falling within Guilford’s concept of "bal
ance . "
The Interview
The purpose of the interview was primarily to ex
plore possible anchorages for role perceptions, as these
might be revealed in the kinds of referents used by the
respondents. In other words, the interview was designed to
be used as an instrument to throw light on the frames of
reference within which the delegates appeared to be per-
4®Guilford, jdjd. cit. , p. 216.
122
ceiving the various categorical roles. Closely related to
this was an attempt to learn what out-of-Council role ex
pectations seemed to be involved in this frame of refer
ence. No statistical analyses of the interview materials
was planned.
By the time the interviews were begun, the prelimin
ary tabulation of the questionnaire data was already in an
advanced stage. An inspection of the se data suggested that
the time variable, around which a number of the hypotheses
had been built, would not prove as significant as had been
believed at the outset of the study. Consequently, an ef
fort was made in the interviews to follow up this early
impression by trying to discover what effects longer dura
tion of Council participation might have upon the percep
tions of various categorical roles. For example, did it
seem possible that one explanation for lower consensus
among long-term participants (in the event that this proved
to be statistically significant) might be because the par
ticipants, being more secure in their Council positions,
felt freer to individualize their responses, rather than to
follow the more characteristic pattern of the role judge
grouping which the participant represented.
Perceptions concerning the place of the layman in
the total Council process were also explored in the inter
views. For example, did it appear that he was perceived by
some as having primarily an extrinsic— rather than intrin
123
sic--value to the total process. In other words, was the
layman being viewed as having something to offer in the
form of an approach to problem solving and a point of view
essential to Council enterprise, or was he accepted as a
necessary part of the social machinery which nominal
democracy imposes upon such enterprises in our culture.
A general interview patter was adopted at the out
set, with effort being made to insure flexibility in the
approach and freedom of expression. Rapport did not appear
to be difficult to establish, especially after the inter
viewee had been assured that the purpose was not to make a
consistency check on the questionnaire responses. Earlier
contact with some of the delegates had revealed this as be
ing a major source of anxiety surrounding the prospect of
being selected for an interview.
How was the interview sample chosen, and what was
the rationale for its size? Bearing in mind the purpose of
the interview— primarily a search for possible role percep
tion anchorages, from which no statistical computations
were anticipated— it was believed that all significant seg
ments of Council delegates, including those who did not
return the questionnaire, should be represented in the sam
ple. Care was taken to guarantee randomization in the se
lection of the subjects, and stratification approximating
the distribution, according to role judge groupings. This
was done by selecting each n'th case from a list of code
124
numbers of role judges by Council tenure. The code numbers
had been assigned to the delegate at the time the question
naire was distributed.
The standard formula for adequacy of sample size
( .3 \f ~ pop. N) demands that no segment of the sample should
have fewer than 3.38 cases. The smallest single category
used in the study was the short-term Non-MSW role judge
group, which included six cases. To follow the above mini
mum here would require not less than three of those six
cases, or a 50 per cent sample, and the value of the inter
view findings to the total study did not justify such ex
tensive coverage (64 interviews). On the other hand, fol
lowing the demands of the above formula only in the broader
categorical groupings, appeared too small a sample for the
purposes of the study. For example, the smallest category
of role judge grouping, irrespective of length of tenure,
was the Non-MSW, which consisted of 22 cases. Interviewing
three cases from this group would be at the rate of 14 per
cent as the total sample. The subjective and unstructured
nature of the interview design seemed to demand a sample
larger than 14 per cent, or a total of eighteen interviews
from the entire delegate body of 127. Consequently, a
compromise was made between 50 per cent, which seemed ex
cessive, and 14 per cent, which was deemed to be insuffi
cient, and an attempt was made for representation which
would be proportional to the total sample. In addition, it
125
was also attempted to make sure that there were no less
than three cases from any one group of non-respondees.
This compromise resulted in a 24 per cent stratified sample
of the total delegate body— seventeen interviews with so
cial worker delegates and twelve with laymen.
Responses to the Questionnaire
The pretested questionnaire was sent to all 127
listed members of the St. Paul Case Work Council in October,
1958. Of these seventy-two, or 57 per cent, were agency
personnel and are referred to in this study as social work
ers, while fifty-five, or 43 per cent, were laymen.
Seventy-three filled out questionnaires, or 58 per cent,
were returned. However, two of these came in too late to
be included in the tabulations and one was incompletely
filled out. The study, therefore, is based on seventy
delegate responses, the categorical distribution of which
is shown in Table 1.
What light can be thrown upon the 18 per cent of the
agency personnel and the 26 per cent of laymen who did not
respond to the questionnaire, but whose names were on the
Council role of 127 delegates? An effort has been made to
analyze a questionnaire mortality pattern in terms of the
types of agencies and organizations from which the respec
tive delegates come to the Council. Most of the delegates
who failed to return the questionnaire were contacted by
126
TABLE 1
DISTRIBUTION OF QUESTIONNAIRE RESPONDENTS,
BY ROLE JUDGE GROUPINGS AND
LENGTH OF COUNCIL TENURE
Length of MSW Non-MSW Lay
Council Role Role Role Total
Tenure Judges Judges Judges
2 Yrs. & Less 16 6 14 36
3 Yrs, & More 9 16 9 34
Total 25 22 23 70
telephone in an effort to encourage them to fill out the
questionnaire, or to learn why they had not done so. The
response data are summarized as follows:
Social
Workers Laymen
Refused...................... 0 5
Promised to return the question
naire but did n o t ..... 8 7
Have resigned from Council, or do
not consider themselves as
member s................. 2 5
Are attached to marginal agencies
or organizations........ 2 3
Out of town, or have left the
city...................... 2 1
Too new in the Council to "know
what it is all about"49 . . . 4 3
Unable to determine........ 5 7
Total 23 31
127
Examining the questionnaire responses and mortality
pattern by type of agency seems to throw additional light
on the matter. If the total Council delegation be categor
ized by the kind of agency represented, the following three
types might be encisioned: Type I, those agencies where
case work services are the total or major concern; Type II,
those which perform some case work services but might be
designated as primarily rendering related services; and
Type III, those welfare-conscious agencies or organizations
which do not perform actual case work or related type ser
vices. If the questionnaire response pattern is examined
in terms of the response potential for Type I and Type II
agency delegates, we find a 69 per cent response, as against
the gross return of 68 per cent. Further, if the response
potential for Type I agencies is viewed alone, there was a
77 per cent response to the questionnaire. Conversely, the
highest single response mortality is among laymen who are
delegates from Type III organizations— those most peripher
al to the Council purpose and content.
The experience of the writer in the Council supports
the notion that the bulk of the work, as well as the major
ity of the concern, are carried by delegates from Type I
^The questionnaire, having been distributed in
October, seemed to justify this response as new delegates
had attended a maximum of only two meetings prior to Octo
ber .
128
agencies, with considerable support from, delegates from
Type II agencies. It would seem reasonable, therefore, to
assume that a fairly representative response has been ob
tained. In fact, the question might well be raised as to
whether a larger response from the low-response categories
might not have introduced a more crucial bias into the
findings of the study than the lack of response may have
done. Certainly, this would be true for those who were new
to the Council and had attended no more than one meeting
at the time the questionnaire was issued.
Summary
The general hypotheses were introduced, together
with their justification in role theory. The research hy
potheses were specified and their relations to the general
hypotheses were shown. Construction of the sections of the
questionnaire was described and the different kinds of data
which each were designed to yield (research hypotheses)
were explained. The method used for the construction of
the modal pattern was presented in detail, as many of the
measurement and comparative efforts of the study are de
pendent upon data which this technique yielded.
The process whereby the questionnaire was pretested
was depicted and the values from the process were shown to
be largely those of language refinement and improved
structural arrangement of the questionnaire. A justifica
tion for the use of nonparametric statistical techniques
for the analysis of the data was offered. This was given
in terms of the nature of the population studied, the size
of the N, and the nature of the data which the questionnaire
produced. A rationale for setting the level of signifi
cance at .05 was based large on the anticipated use to
which the data would likely be put. The purpose for incor
porating interviews in the study design was explained as
providing cues to reference-group anchorages for role per
ceptions. The method used for selection of the interview
sample was detailed. Results of an inquiry into the rea
sons for the mortality of the questionnaire were presented
in defense of the size of the N used for the study.
CHAPTER V
FINDINGS OF THE STUDY
Role Consensus by Length of Council Tenure
As stated in Chapter IV, the three instruments con
tained in the Role Consensus Battery represent three kinds
of approaches to the same problem, which was to determine
the degree of role perception consensus among various role
judge groupings of the Council delegates.^ The three dif
ferent approaches to this problem were discussed in detail
in Chapter IV. In designing the instruments within the
It must be remembered that each respondent is
asked, at varying points throughout the questionnaire, to
indicate his perception of each of the following: his own
(personal) role, his own categorical role, and the two re
lated roles. The two latter perceptions are concerned, not
with the respondent’s direct perception of the categorical
roles as such, but rather with his estimate of the percep
tions which the role incumbents themselves have of their
respective roles. In a sense, then, these might be said to
be "indirect" perceptions. They involve the respondent’s
"taking the role of the other." This holds true for almost
the entire study, not just the section on the Role Consen
sus Battery.
Since it becomes extremely verbose, as well as po
tentially confusing, to fully repeat the nature of these
indirect perceptions each time they are dealt with in the
analysis, a symbolic form of representation has been adopt
ed. Therefore, whenever these indirect perceptions are
being referred to, the letters REOP (respondent's estimate
of others' perceptions) will be used throughout the analy
sis of the data.
130
131
Battery, the assumption was made that the results of all
three would prove mutually supportive. All were designed
to test Research Hypothesis 1.
Research Hypothesis 1
"Role consensus is greater among long-term Council
participants than among short-term participants."2
Instrument B.— Does the total sample (irrespective
of any categorical breakdown by role judge grouping) reveal
greater consensus with reference to long-term participants
than with short-term participants, as they view (REOP) each
q
of the three categorical roles? The Sign Test results0 , as
shown in Table 2, indicate that this hypothesis is not sup
ported at the level of significance adopted for the study,
namely, ,0b. (See Table 2, "Total sample.")
Further, the hypothesis is likewise not supported
when the time variable is applied to the categorical break
down by role judge groupings (MSW, Non-MSW and Lay), as
they perceive (REOP) the three categorical roles (Social
Worker, Layman, and Staff). (See Table 2.)
Inasmuch as Instrument B is composed of 14 role-cue
^Terms used in the hypotheses have been defined in
Chapter I. As "long-term" refers to those delegates who
had been in the Council three or more years, the symbol 3+
is used to refer to such participants. The symbol 2- is
used to refer to "short-term" participants who had been in
the Council two years or less.
%or an explanation of the steps used in computing
the Sign Test, see pp. 114-116.
132
TABLE 2
SUMMARY OF THE PROBABILITIES ASSOCIATED WITH THE
OCCURRENCE OF VALUES REGARDING ROLE CONSENSUS
AND LENGTH OF COUNCIL TENURE, WHEN EACH OF
THE CATEGORICAL ROLES IS PERCEIVED
BY ROLE JUDGE GROUPINGS
Role
Judge
Grouping
Categorical Role Perceived
Social Work Lay Staff
Total sample . 090 .605 . 212
MSW .500 . 5u0a . 274a
Non-MSW .291 . 194 .212
Lay .212 .090 .395
aThese are the only two cases where the probability
figures reflect a majority of plus values. In all other
instances, the figures are indicative of a majority of mi
nus values, which is in a direction contrary to the hypoth
esis.
statements, it would have been necessary for the data to
have yielded a minimum of 11 plus values ^indicating more
consensus among the long-term participants) as against no
more than three minus values (indicating more consensus
among the short-term participants), for the hypothesis to
have been supported at the .05 level of significance (using
the Sign Test technique). This, of course, is a very
stringent standard to apply to such data.
It is interesting to note, moreover, that when the
Sign Test technique is applied here, it reveals a consistent
133
tendency for the minus values to predominate (in 9 out of
12 cases). This is in precisely the opposite direction
from the hypothesis, i.e., suggests that perhaps 2-> 3+.
The only exception to this is when the MSW view (REOP) the
Lay and Staff roles. The possible implications of this re
verse tendency, although it is not statistically signifi
cant, will be discussed later, in the light of the findings
from Instruments D and E.
Instrument D.--When the various categorical roles
are perceived (REOP) by the entire sample, irrespective of
role judge groupings, the hypothesis is not supported at
the .Ob level of significance with Instrument D, any more
than it was with Instrument B.
IVhen the participants are divided by role judge
groupings, the same situation prevails, with the one excep
tion that long-term laymen do show more consensus than
short-term laymen (.046, significant at the .Ob level).
Instrument E.— What does the total sample reveal,
regarding consensus (3*}>2-), when viewing the respective
categorical roles in terms of which one should assume the
"most" and "least" responsibility for initiatory action?
Sign Test results here again indicate that the hypothesis
is not supported at the ,0b level of significance. It is
likewise not supported when it is examined in terms of the
role judge groupings. None of the groupings of role judges
134
show the factor of longer tenure in the Council to be a
significant determiner of increased consensus, when viewing
any of the categorical roles.
As was found in the analysis of Instrument B, in
spection of the plus and minus values shows that there Is a
consistent tendency for the minus values to predominate (in
7 out of 8 cases), when perceptions are being made in the
light of the "most responsible" and "least responsible"
roles. This again suggests a possibility that perhaps
2-> 3+. The only exception to this is when the MSV/ make
their selection of the "most responsible" role (.151, sig
nificant at the .05 level).
Some possible interpretations of the battery re
sults .— The tendency for the data to require the rejection
of Research Hypothesis 1 (at the .05 level) in the case of
all three of the instruments, might be explained by one or
a combination of the following factors:
1. The three instruments, as constructed, may not
be sufficiently refined to reveal the smaller and more
subtle differences, i.e., they may be capable only of show
ing general trends.
2. The power-efficiency limitations of the Sign Test
technique, particularly if they are coupled with #1 above,
may have resulted in a failure to reveal differences actu
ally inherent in the data, i.e., they may be forcing the
135
discarding of some of the more subtle, but potentially
important, differences which actually exist.^
3. There may be no actual difference in the degree
of role consensus, as the length of tenure varies, in this
particular group setting.
4. The direction of the difference, being consist
ently opposite to the hypothesis, may indicate that, given
a larger N, a hypothesis stating that 2-> 3-v might well be
supported to a statistically significant degree.
Since the above mentioned reverse tendency appeared
consistently in all three instruments in the battery, it
would seem reasonable to assume that some factor other than
chance is operating with regard to the time variable and
role consensus. Yet the data will not permit us to assert
any statistically reliable conclusions as to the meaning of
this tendency.
It may be relevant to ask why there was support for
the hypothesis among the laymen on Instrument D. It would
appear, first of all, that this instrument ("Who Said
That?") is the least structured of the three, and if this
is true, it would require that finer, and more subtle,
distinctions be made in responding to it. In addition,
4See Robert K. Merton, George G. Reader, and Patri
cia L. Kendall, The Student Physician: Introductory Studies
in the Sociology of h/lediical Education (Cambridge: Harvard
University JPress , 19^), pp. 303-i3t)4.
136
the Lay role seems to be the least structured of the three
categorical roles in the Council, and thus it is not un
reasonable to assume that the laymen would have the great
est difficulty in attaching these role-cue statements to
the appropriate roles. Consequently, their apparent devia
tion from the general pattern set by the other role judges
might, quite naturally, be expected to be greater than the
deviation among the incumbents of the more structured
roles.
Role Consensus by Role Judge Categories
Research Hypothesis 2a
"Role consensus is greater among social workers than
among laymen. 1 1 (SW^>L)
On the basis of the rankings of the potential modal
scores for each of the role judge categories, it can be
seen in Table 3 that this hypothesis is strongly supported.
With only one exception— that of laymen viewing their own
role— »the social workers consistently appear at the top of
the rank order distribution of scores and the laymen at the
bottom.
Research Hypothesis 2b
"Role consensus is greater among Master’s of Social
Work degree participants than among those without such
a degree." (MSW> Non-MSW)
Table 3 shows this hypothesis is likewise strongly
supported. When MSW participants view (REOP) other cate-
137
TABLE 3
RANK ORDER DISTRIBUTION OF POTENTIAL CONSENSUS
SCORES OF ROLE JUDGE GROUPINGS, AS THEY
PERCEIVE EACH OF THE CATEGORICAL ROLES
AND THEIR OWN (PERSONAL) ROLES
« - i - r . r, n • Potential
Role Judge Role Being Consensus
Grouping Perceived Score
MSW
Soc. Work. 4b
MSW Own 44
MSW Layman 40
Non-MSW S oc . W or k. 40
Non-MSW
Lay 40
Non-MSW Staff 40
Lay Own 38
Non-MSW Own 37
MSW Staff 36
Lay Soc. Work. 34
Lay Lay 34
Lay Staff 32
gorical roles and their own (personal) role, they appear at
the top of the rank order distribution, the only one ex
ception being when they view (REOP) Staff.
138
Goodness of Role Perception
Research Hypothesis 3
"Good role percedvers (those nearest the modal
pattern) in terms of their own categorical roles tend
to be good role perceivers of counter roles."
Good role perceivers, as defined in this study,
are those role judges whose role perceptions approximate
the modal pattern for their own respective categorical
roles. The index of good role perception is that score
resulting from the summation of the weights of the indivi
dual role-cue statements in Instrument B, i.e., the higher
the score, the better the role perception. The application
of this notion of good role perception to the data involved
the following steps:
1. A modal score for each respondent was com
puted. This was done by comparing his item responses with
the modal pattern for his own categorical role.
2. These individual scores were then grouped ac
cording to role judge categories.
3. Through the use of the Spearman Rank Correla
tion Coefficient (Rho) technique, a comparison was made
within each role judge category to determine the extent of
role perception consistency, vis-a-vis the hypothesis.
The above process yields data which supports
Research Hypothesis 3 at the .01 level of significance,
which is beyond the level set for the study (.05). Table 4
139
summarizes these data and indicates that, with only one ex
ception (MSW viewing the Lay role), the findings are sup
ported at the .05 level.
TABLE 4
SUMMARY OF CRITICAL VALUES OF RHO DEPICTING
THE DEGREE OF CONSISTENCY IN
THE PERCEPTION OF ALL CATEGORICAL ROLES,
BY ROLE JUDGE GROUPINGS
Role
Judge
Grouping
No.
of
Case s
Categorical Roles Perceived
Social
Work
Lay Staff Own
MSW 25 — . 296a .544 .828
Non-MSW 22 - .644 .630 .642
Lay 23 .617 - .680 .681
aNot significant at the .05 level. Moreover, since
this case involved the greatest proportion of tied ranks,
the Rho correction for ties was applied here. No signifi
cant difference was found, however, in the resultant crit
ical values. The corrected critical value of Rho was .290.
Although the critical values of Rho, as seen in
Table 4, are statistically significant (except one), it is
a fact that they are not necessarily indicative of good role
perception in all cases. What they do show, rather, is that
good role perceivers of their own categorical role tend to
be good role perceivers of related roles; while poor role
perceivers of their own categorical role tend to be poor
role perceivers of related roles. They indicate, then, a
140
consistency in the type of role perception.
An additional test of the "goodness" of role per
ception was applied to the data by utilizing the modal
scores derived from Instrument E. Here the underlying gen
eralization was made that there would be a consistency of
good role perception, when the respondents assigned the
"most" and the "least" responsibility for initiatory action.
Rho was again used to determine the correlation between the
"most" and "least" modal scores, for each of the role judge
groupings.
A statistically significant relationship was found
between the two rankings (at the ,0b level) for both the
MSW (Rho = .34) and Lay (Rho r .48) role judge categories.
Such a significant relationship was not found for the Non-
MSW category (Rho » .27). However, since approximately 70
per cent of the total sample is accounted for in the stat
istically significant groupings (MSW and Lay), it would
appear that there is a considerable degree of consistency
in the goodness of role perception. In addition, it should
perhaps be mentioned that the results of this test seem to
indicate that the use of the "least" scores in the analysis
of Instrument E is justifiable.^
^A question had arisen in the mind of the writer, at
the time the questionnaire data were being tabulated, as to
the validity of the data in the "least" column, and hence
their ability to yield valid scores. An examination of the
data seemed to suggest that perhaps the respondents had
141
Role Clarity
Research Hypothesis 4a
"Role clarity is greater among long-term partici
pants than among short-term participants." (3*. > 2- )
The Chi Square test is the basic statistical measure
utilized throughout this part of the analysis. It is sup
plemented by percentage distributions, however, when these
appear to help reveal patterns of consistency which exist
but which are not statistically significant at the .05
level, and which might be suggestive for future research.
When the various categorical roles (and own) are
perceived (REOP) by the entire sample, irrespective of role
judge groupings, the hypothesis is not supported at the .05
level of significance. (See Table 5, and Appendix C,
Table 3-C. )
Inspection of the data reveals that further analysis
is necessary. When all of the delegates are viewing (REOP)
been less discriminating in their judgments in the "least"
column than in the "most" column. It was felt that there
might be two possible reasons for this. First, having se
lected the "most" important role, the step of selecting the
"least" meant, in effect, that all three of the roles were
being given a rank order placement. This might have been
asking for a greater refinement than many respondents were
able or willing to make. Second, having selected the
"most" important role, it was thought that less time and
effort were likely to be expended in selecting the "least"
important role, particularly in the light of the large
amount of time which the filling out of the questionnaire
required. However, the data here does not support this
suspicion.
142
TABLE 5
SUMMARY OF CHI SQUARE CRITICAL VALUES AND
ASSOCIATED PROBABILITIES SHOWING THE EXTENT
OF INDEPENDENCE OF STATED ROLE CLARITY
AND LENGTH OF COUNCIL TENURE,
WHEN EACH COUNCIL ROLE IS
PERCEIVED BY ALL DELEGATES
Council
Role
Perce ived
No. of
Delegates
Perce iving
Chi Square
Critical
Values
Degree s
of
Freedom
Pr obabil-
itya
Social Worker
68 c
2.27 2 > .30 < .60
Staffb
- - - -
Layman 68° 2.00 2 > .30 <; .50
Own 70 1.94 2 y .30< .50
aCare must be taken in interpreting the apparent
independence of these variables, as indicated by the low
probabilities. In the case of the Lay and Own roles, the
percentage distributions indicate that there is a trend in
the direction of the hypothesis, but not to a statistically
significant degree. In the case of the Social Worker role,
however, the low probability is a reflection of a highly
skewed distribution, favoring clarity, regardless of Coun
cil tenure.
^The distribution was here so skewed in the direc
tion of clarity that the analysis was not only unneces
sary, but impossible, to compute owing to small cell values
in all but the "unusually clear" cells.
cReduced N due to two uncompleted questionnaires.
the Social Worker role, we find that regardless of the time
variable, this role is perceived as being more clear than
unclear (91 per cent among the 2- participants, and 8b per
cent among the 3f).^ Of those viewing the Social Worker
role in this manner, however, only 37 per cent of the
143
short-term participants perceive (REOP) it as being "usual
ly clear," whereas 49 per cent of the long-term partici
pants perceive it as such. This fact suggests that, per
centage-wise, the data tend to move in the direction of the
hypothesis (3+ >2-).
Similarly, it is apparent when the Staff role is
being viewed (REOP) that, regardless of the time variable,
this role is perceived as being overwhelmingly more clear
than unclear (100 per cent among the 2- participants, and
94 per cent among the 34-). The fact that 77 per cent of
the total sample view it (REOP) as being "usually clear"
seems to demonstrate the high degree to which the role is
structured.
When role judges perceive their own (personal) role,
61 per cent of the short-term participants and 76 per cent
of the long-term participants state that they perceive this
role as being more clear than unclear. From the percentage
point of view, this again tends to support the directional-
Although the questionnaire provided for four de
grees of stated clarity, it was found that the data could
be treated meaningfully within a dichotomy, i.e., percep
tion of a role being more often clear than unclear, or
more often unclear than clear. This dichotomy will thus
be utilized throughout the following discussion of role
clarity. In those instances where the degree of clarity or
lack of clarity is found to be extreme, this fact will be
indicated by the use of the terms "usually clear" or usu
ally unclear," accompanied by the appropriate percentage
figure. See Appendix C, Table 1-C, for percentage distrib
utions .
144
ity specified in the hypothesis.
In contrast to the above-stated tendency toward role
clarity (with regard to the Social Worker and Staff roles),
the Lay role is perceived (REOP) by 46 per cent of the
total sample as being more unclear than clear. This would
appear to support the belief, stated earlier, that the Lay
role is the least structured of the three categorical
roles. Although not statistically significant at the ,0b
level, we find that even here there is a percentage differ
ential by time which favors Research Hypothesis 4a, i.e.,
49 per cent of the short-term participants perceive (REOP)
the Lay role as being clear, and 51 per cent as unclear;
whereas 61 per cent of the long-term participants perceive
it as clear, and only 39 per cent as unclear.
Hypothesis 4b
"Role clarity is greater among social workers than
among laymen."
Regardless of which of the categorical roles is be
ing perceived (REOP), this hypothesis is not supported at
the .05 level of significance. (See Table 6, and Appendix
C, Table 4-C.)
Here again, further analysis is necessary to reveal
a complete picture of the data. Such an analysis shows a
role perception pattern which is strikingly similar to that
which appeared in the case of Research Hypothesis 4a above.
Regardless of the role judge grouping, the Social
145
TABLE 6
SUMMARY OF CHI SQUARE CRITICAL VALUES AND
ASSOCIATED PROBABILITIES SHOWING THE EXTENT
OF INDEPENDENCE BETWEEN STATED ROLE CLARITY
AND TYPE OF ROLE JUDGE GROUPING,a
WHEN EACH COUNCIL ROLE IS
PERCEIVED BY ALL DELEGATES
Council
Role
Perceived
No. of
Delegates
Perce iving
Chi Square
Critical
Values
Degree s
of
Freedom
Probabil
ity
Social Worker 68c 1.87 2 >.30 < .50
Staff& - - - -
Layman 63c 1.64 2 >.30< .50
Own 70 1. 01 2 >.30 < .50
aRole judge groupings were based on differences in
the amount and type of professional training.
^See footnote '*b* * to Table 5.
cReduced N due to two uncompleted questionnaires.
Worker role is viewed (REOP) as being more often clear than
unclear (96 per cent among the laymen, and 84 per cent
among the social workers. (See Appendix C, Table 2-C. )
The Staff role is likewise viewed (REOP) as being
more clear than unclear (100 per cent among the laymen, and
96 per cent among the social workers. Moreover, we again
find, as was true with Research Hypothesis 4a, that 77 per
cent of the total sample view this role (REOP) as being
"usually clear"— the highest degree possible on the clarity
scale. These figures would all seem to give additional
146
support to the earlier conclusion that the Staff role is
the most highly structured of the three categorical roles.
Although not statistically significant at the .05
level, the percentage distribution would appear to support
the directionality of the hypothesis, when the respondent's
own (personal) role is being perceived. Among the laymen,
61 per cent view their own role as being more often clear
than unclear, while 72 per cent of the social workers view
their own role in this manner.
The consistency of the perception pattern continues
when the Lay role is viewed (REOP). Once again this role
is perceived (REOP) by 46 per cent of the total sample as
being more unclear than clear. It should be pointed out
that this is identical with the percentage which was found
above when the Lay role response was examined with regard
to the time variable. Consequently, this finding further
supports the previous data which indicate that the Lay
role is the least structured of all the Council categorical
roles. In addition, as was true in the case of Research
Hypothesis 4a, there is a percentage differential by role
judge grouping which moves in the direction of the hypoth
esis. Among the laymen, 57 per cent perceive (REOP) the
Lay role as being more often unclear than clear; while among
the social workers, the converse is true, i.e., it is per
ceived (REOP) by 60 per cent as being more often clear than
unclear.
Role Involvement
147
Research Hypothesis 5a
"Role involvement is greater among long-term partic
ipants than among short-term participants." (7* >2- )
The analysis of this part of the data is also based
on the Chi Square technique, supplemented by a percentage
breakdown whenever it appears that such a breakdown might
be useful in providing a more complete picture of the data.
The application of the Chi Square test to an analy
sis of the entire sample reveals that the hypothesis is not
supported at the .Ob level of significance. In fact, it
would appear here that there is no relationship between
role involvement scores and length of Council tenure.
(x^ = .Ob, 1 df, p >.80 < .90. ) Similarly, when we view the
data by role judge categories, the hypothesis is likewise
not supported at the .Ob level.
Research Hypothesis 5b
"Role involvement is greater among social workers
than among laymen."
An examination of Table 7 indicates that the social
worker role judge category is not homogeneous with regard
to the variables being treated in this hypothesis. This is
exemplified by the fact there is a substantial differ
ence between the MSW and the non-MSW grouping, when role
involvement scores are examined. The MSW grouping, tends
to move in the direction of the hypothesis, but the non-MSW
148
TABLE 7
ail SQUARE TABLE SHOWING THE EXTENT OF
INDEPENDENCE OF ROLE INVOLVEMENT SCORES
AND TYPES OF ROLE JUDGE GROUPINGS
Involvement
Scores
Role Judge Grouping
Total
MSW Non-MSW Lay
0 - 1 7 10 13 13 36
18 - 37 15 9 10 34
Total 25 22 23 70
Chi Square ■ 2.12, (p > .30 < .50)
are found to approximate more closely the lay pattern,
i.e., that of less stated role involvement.
Hypothesis 6a
"Social Workers tend to anchor their role perceptions
of the Social Worker and Staff categorical roles more
within the professional training orientation, whereas
laymen anchor their perceptions of the lay categorical
role more within the broader societal orientation."
The analysis of Council role anchorages (Section F-3
of the questionnaire) is based on a ranking of those three
sources of influence, out of a possible eight, which are
perceived by the respondents as being the most important
in influencing the performance of their own (personal)
Council role; and rankings of those perceived (REOP) as
most important in the performance of each of the three
149
Cj
categorical roles.
Tabulation of the data revealed that only six out of
the eight potential sources of role influence had been se
lected by the respondents as being among those three most
important to the performance of the various council roles
(items "g" and "h" were consistently omitted). The six
items which had been selected, however, were grouped by
their general content into three basic types of orienta
tion; "Professional Training" (items "b" and "d"), "Broad
er Societal" (terns "a" and "c"), and "Intra-Council" (items
"e" and "f"). Analysis of the data, within the framework
of these three types of orientation, yielded results in
support of Research Hypothesis 6a (see Table 8).
It is quite apparent that social workers do anchor
their perceptions of both the Social Worker and Staff cate
gorical roles (REOP), as well as their own (personal)
roles, in the professional syndrome. Furthermore, item
"b" consistently appears to be first in importance in in
fluencing the performance of these Council roles.
Laymen, on the other hand, anchor their perceptions
of the Lay categorical role (REOP) in the broader societal
^The rationale for including only the first three
choices in this analysis is supported by the high concen
tration of total potential selections found among these
first three, regardless of the role judge grouping, or the
categorical roles being perceived. This concentration is
never less than 53 per cent of the total potential, and it
is usually over 65 per cent.
TABLE 8
RANKING OF THE THREE SOURCES OF INFLUENCE PERCEIVED AS BEING MOST IMPORTANT
TO THE PERFORMANCE OF EACH CATEGORICAL ROLE, BY ROLE JUDGE GROUPINGS3
« = » - — ■ .in -■ ■ I i .m w u i ^ i I I g a g - a * ^ m a m - ■ — — ■ —..." gj n-^iuii ,n — « . -1 .^.1 —11
Possible*3 Orien-c MSV/ Judges Viewing Non-MSW Judges Viewing Lay Judges Viewing
Sources tation Each Categorical Role Each Categorical Role Each Categorical Role
Influence
o y 1
drome
Own Soc.
Work
Staff Lay
Own Soc.
Work
Staff Lay Own Soc.
Work
Staff Lay
"a" Soc. 3 1 3 2 3 1 2 1
"b” Prof. 1 1 1 1 1 1 3 1 1
«c" Soc. 2 2 3
"d" Prof.
3 2 3 2 3
"e" Coun. 2 3 2 1 2 3 2 2 1 3 2 2
i i f T i
Coun. 3 3
aThe numbers in body of table refer to the rank order of the three most important
weighted choices for each categorical role. (Duplicate figures indicate tied ranks. )
^These letters correspond with the items in Section F-3 of the questionnaire.
(See Appendix A. )
Abbreviations refer to the three basic types of orientation: Broader Societal
(Soc. ), Professional Training (Prof.), and Intra-Council (Coun.). See p. 149.
150
151
syndrome. However, when they view their own (personal)
role, the intra-Council orientation becomes the more impor
tant, and the broader societal orientation second in impor
tance .
It is highly significant that, when viewing role
anchorages, all role judges perceive the incumbents of the
various categorical roles (REOP) in the same manner in
which these incumbents perceive themselves.
Although it was originally written for a slightly
different purpose, there are data from Section F-l (Appen
dix A, p. 216) which reinforce the above findings concern
ing role anchorages. In this Section, the respondent was
asked to indicate to what degree of consideration he would
assign each of eight possible reasons for his willingness
to spend time in Council activities. It was possible to
analyze most of these reasons within the framework of ori
entation adopted for Section F-3.
A percentage distribution of the responses to this
Section reveals that social workers tend to select the pro
fessionally oriented reasons (a. "The kind of society in
which we live and the way we are expected to do things in
our society," and f. "My observations of the way things are
done in the Council.") as being the major ones in their
willingness to participate in Council activities. Laymen,
on the other hand, tend to give prime consideration to
reasons which have a broader societal orientation (d. "In
152
general the way things are done in my profession or type of
work."), or which deal more with personal satisfaction (c.
"The way my friends and respected associates tend to look
at such matters."). Moreover, laymen seem to attach the
least importance to those professionally oriented items
which appear most important to the social workers. This
would seem to suggest that the layman's training, with re
gard to his particular professional role, does not include
expectations which presume his participation in community
activities of this nature.
Still another approach to the study of role anchor
ages is through an examination of the perceived importance
of the Council role to some of the respondent's valued out-
of-Council roles. Sections F-5 and F-6 (Appendix A, pp.
219 and 220) were constructed with this in mind. Section
F-5 asks the respondent to indicate the relative values
which he perceives his Council contacts to have in each of
the following out-of-Council situations: a. "personal
friendships," b. "business, occupation, or profession," or
c. "getting ahead in the community." Section F-6 asks the
respondent to classify his friends and associates in terms
of which ones he "would like to have know" about his Coun
cil activities.
Application of the Chi Square technique to the an
swers from Section F-5 yielded statistically significant
support for the high degree of professional orientation
153
which has already been noted among social workers as con
trasted with laymen (x^ a 8.27, 2 df, p >.01 <.02). (See
Appendix C, Table 5-C. ) It showed no statistically sig
nificant differences between social workers and laymen,
however, when they considered the importance of their Coun
cil contacts in helping them "get ahead in the community,"
or in "forming personal friendships" outside of the Coun
cil.
When we examine the answers in Section F-6, we find
here, again, that social workers, when choosing among those
friends and associates whom they would "like to have know"
about their Council activities, consistently select items
which are professional oriented ("a" and "c" with strong
support from "d"). Laymen, on the other hand, although
they select a wider range of items, tend to be consistent
in selecting a cluster around a non-professional friend-
associate nucleus. It is interesting to note that both
social workers and laymen indicate very few instances where
they would prefer to have their Council activities "not
known" to certain of their friends or associates outside of
the Council.
Research Hypothesis 6b
"Those with Master of Social Work degrees are more
professional training oriented than are those social
workers without such a degree, when perceiving the
Social Worker and Staff categorical roles."
Further inspection of Table 8 (page 150) reveals
154
that this hypothesis is also supported. Among the first
three choices made by the MSI// role judges, the professional
training syndrome (items b. "My professional education and/
or training," and d. "In general the way things are done in
my profession or type of work.") consistently appears, when
viewing the Social Worker and Staff categorical roles
(REOP), and their own (personal) role.
Non-MSW role judges, on the other hand, are not as
consistently professional-training oriented, in their role
anchorages, as are the MSW. While professional training
(item "b") is consistently their first choice, they deviate
from the MSW grouping in that the intra-Council (item "e")
and the broader societal (item "a") orientations are ranked
as being second and third in importance.
In view of both the total number of possible choices
and the freedom to rate any item as first, second, or
third in importance, the fact that there was complete unan
imity among the respondents on the first choice, for all of
the categorical roles, is highly significant.
Research Hypothesis 6c
"Long-term participants are more intra-Council ori
ented in their role perception anchorages than are
short-term participants."
It should be recalled that whenever the time vari
able has been considered in this study, it has not appeared
as a significant factor in Council role perception.
155
Table 9 presents additional findings which are con
sistent with this pattern. Complete agreement is shown on
the most important role anchorages for each of the categor
ical roles, regardless of the length of Council tenure.
TABLE 9
RANKING OF THE THREE SOURCES OF INFLUENCE
PERCEIVED AS BEING MOST IMPORTANT TO
THE PERFORMANCE OF EACH CATEGORICAL ROLE,
BY LENGTH OF COUNCIL TENURE3
Possible
Sources
of In-
fluence
Or ie n-
tation
Syn
drome
Council Tenure
2 yrs. or less
Council
3 yr s.
Tenure
or more
Own
Soc.
Work
Staff Lay Own
Soc.
Work
Staff Lay
"a" Soc. 3 2 1 3 3 1
"b" Prof . 1 1 1 1 1 1
i t c i ' i
Soc. 2 2
"d" Prof. 2 2 3
"e" Coun. 2 3 1 3 2 2 2 1
I I £11
Coun. 3 3
aThe numbers in the body of the table refer to the
rank order of the three most important weighted choices for
each categorical role. (Duplicate figures in a column in
dicate tied ranks.)
bThe se letters correspond with the items in Section
F-3 of the questionnaire. (See Appendix A.)
cThe abbreviations used here refer to the three
basic types of orientation; Broader Societal (Soc.), Pro
fessional Training (Prof.), and Intra-Council (Coun.).
See page 149.
*S5
156
Item "b" is consistently chosen as being the most
important anchorage for the Social Worker and Staff categor
ical roles, while item "a" is consistently the first choice
for the Lay role. Moreover, item "c" (the other item in
the broader societal syndrome) appears among the first
three choices for both long-term and short-term partici
pants, when the Lay role is being perceived.
The data from Section F-5 further support the ap
parent lack of significance of the time variable in Council
role perception. In this case, the Chi Square results
show no statistically significant difference in the per
ceived importance of the Council role to out-of-Council con
tacts and relationships, as length of Council tenure var
ies (x^ « .32, 2 df, p^ .80 ^.90). (See Appendix C, Table
6-C. )
Some Selected Variables Re-examined in
D i f f e r e n t Co mb i na tIons
Research Hypothesis 7a
"The greater the degree of role involvement, the
greater the degree of stated role clarity."
Here, and in subsequent sections, role involvement
scores are treated dichotomously, i.e., those below the
median (0 - 17) and those above the median (18-37). In
all cases, Chi Square was used to measure the independence
of the variables under consideration.
The above hypothesis regarding role involvement and
157
clarity is not supported at the .05 level of significance,
when the Social Worker and Staff categorical roles are be
ing perceived (REOP) by the entire sample.6 It is sup
ported, however, when the Lay role is being perceived
(x2 - 5.93, 1 df, p >.01< .02).
TABLE 10
CHI SQUARE TABLE SHOWING THE EXTENT OF
INDEPENDENCE OF ROLE INVOLVEMENT SCORES
AND STATED ROLE CLARITY,
WHEN THE LAY ROLE IS PERCEIVED
Role
Involvement
Score s
Stated Role Clarity
T otal
More Clear
Than Unclear
More Unclear
Than Clear
0 - 1 7 14 21 35
18 - 37 23 10 33
T otal 37 31 68
Chi Square a 5.93, p >.01< .02.
An inspection of the percentage distribution re
veals that, irrespective of the degree of involvement of
the role judges, the Social Worker role is generally per-
^The Chi Square test yielded a figure of .52 for the
Social Worker role. With 2 df, the resultant probability
was p > .70 .80. The test was not applied to the Staff
role, however. For an explanation as to why, see footnote
"b", Table 5, p. 142.
158
ceived (REOP) as being more clear than unclear (86 per cent
among the low involvement participants, and 91 per cent
among those with high involvement scores). In fact, 77 per
cent of the total sample view the Social Worker role (REOP)
in this manner.
Similarly, it is apparent that the Staff role is
generally perceived (REOP) as being more clear than un
clear, regardless of the scores which the role judges re
ceive on the involvement scale (94 per cent among the low
involvement participants, and 100 per cent among the high
involvement). In this case, moreover, we find that 97 per
cent of the total sample view the Staff role (REOP) in such
a manner, with 77 per cent viewing it (REOP) as "usually
clear"— the highest degree of clarity possible on the
scale. This fact further supports the contention that the
Staff role is the most highly structured role in the Coun
cil .
As was stated earlier, when the Lay role is per
ceived (REOP) we find a departure from the general response
pattern which has been emerging with regard to the Social
Worker and Staff categorical roles. In the case of the Lay
role, the hypothesis under consideration here was supported
at the .05 level of significance. It was found that 60 per
cent of the low involvement delegates perceive the role
(REOP) as being more often unclear than clear, whereas 70
per cent of the high involvement delegates perceive it
159
(REOP) as being more clear than unclear.
Attention should once more be called to the impli
cations which findings such as this have in an analysis of
the manner in which the three categorical roles appear to
be structured. Here, again, as was suggested in the anal
ysis of role clarity, the Lay role would appear to be the
least structured of the three categorical roles.
Research Hypothesis 7b
"The greater the degree of role involvement, the
greater the degree of role consensus.M
Regardless of which of the three categorical roles
is being perceived (REOP), there does not appear to be a
statistically significant relationship between role in
volvement and role consensus, the latter being measured in
terms of the modal scores derived from Instrument B. The
Chi Square values and associated probabilities for the
various roles are summarized in Table 11. (See Appendix C,
Table 7-C, for numbers.)
Perhaps some explanation is in order for the extreme
probability shown regarding the Staff role. In reality,
this figure merely reflects the fact that the consensus
scores for this role were high, regardless of the degree of
role involvement. We find, for example, that although the
involvement scores were widely distributed over a range
extending from two to thirty-seven points, the total sample
had consensus scores which were concentrated as follows:
160
TABLE 11
SUMMARY OF CHI SQUARE CRITICAL VALUES
AND ASSOCIATED PROBABILITIES SHOWING
THE EXTENT OF INDEPENDENCE OF
MODAL CONSENSUS SCORES AND
ROLE INVOLVEMENT SCORES,
WHEN EACH OF THE
CATEGORICAL ROLES IS PERCEIVED
Categorica1
Role
Perceived
No. of Role
Judges
Perceiving
Chi Square
Cr itica 1
Values
Degrees
of
Freedom
Probability
Social Worker 70
2.59 2 >.20< .30
Staff 70 .13 2 ).9G< .95
Layma n 70 3.16 2 >.20< .30
56 per cent in the upper one-fifth of the consensus-score
range, and 81 per cent in the upper one-third.
Research Hypothesis 7c
"The greater the degree of role involvement, the
greater the degree of stated Council role satisfaction."
It should be recalled that role satisfaction, as
treated in this study, is based on a comparison between the
amount of satisfaction felt to be derived from the perform
ance of the Council role, and the amount derived from the
respondent’s most satisfying civic role.
Application of the Chi Square test yielded results
which do not support the hypothesis at the .05 level of
significance (x^ « 2.45, 1 df, p>.10< .20). (See Appendix
C, Table 8-c.) A percentage distribution, however, shows
161
some differences in satisfaction as role involvement in
creases. Only 17 per cent of those with low involvement
scores state that their Council role satisfactions are
greater than the satisfactions derived from their most sat
isfying civic role; whereas 33 per cent of those with high
involvement scores stated this to be the case.
Research Hypothesis 7d
•'The greater the degree of role involvement, the
greater the degree of stated Council role interest."
As was the case with role satisfaction, the amount
of Council role interest is also treated comparatively,
i.e., in relation to the respondents most interesting
civic role.
No statistically significant relationship was found
between role involvement and stated Council role interest
(x^= 1.11, ldf, p > . 20 < . 30). (See Appendix C, Table
9-C.) It is interesting to note, however, that only 33 per
cent of those with low involvement scores state their Coun
cil role interest to be more than their most interesting
civic role; whereas 47 per cent of those with high involve
ment scores make such a statement.
Research Hypothesis 8
"The greater the degree of role consensus, the great
er the degree of stated role clarity."
Regardless of which of the three categorical roles
is being perceived (REOP), there does not appear to be a
162
statistically significant relationship between role consen
sus and stated role clarity.
Again, however, it should be pointed out that the
apparent difference in role structure should be taken into
consideration when interpreting these results. For exam
ple, when the Social Worker role is being perceived (REOP)
by the entire sample, 8b per cent view it as being more
clear than unclear. It would thus seem that the role being
perceived is, in this case, so definite and well struc
tured, that it can be perceived equally well by all of the
respondents, regardless of how they may differ from one
another with respect to other variables.
The highly structured nature of the Staff role is
likewise indicated by the fact that 97 per cent of the
total sample view it (REOP) as being more often clear than
unclear. Moreover, 77 per cent perceive it as being "usu
ally clear"— the highest degree of clarity on the scale.
It is highly significant that these two percentage figures
are identical with those regarding the Staff role under
Research Hypothesis 7a.
While the extreme skewness of the distribution ac
counts, then, for the apparent lack of relationship when
the Social Worker and Staff categorical roles are being
perceived (REOP)^ the probability for the Lay role reflects
"^The Chi Square test yielded a figure of 1.95 for
the Social Worker role. With 2 df, the resultant probabil-
163
a genuine lack of any significant relationship between
these two variables (x^ « .49, 1 df, .70< .80). (See
Appendix C, Table 10-C.) By way of illustration, the per
centage distribution shows, in this case, that almost one-
half of the total sample (4b per cent) perceive this role
as being more unclear than clear. In addition, this is
true regardless of the respondent’s consensus score.
Conceptualizations of Community Organization
as Possible "AnchoVaqe s' fror "Role Perception
Although an item analysis of the instruments con
structed for this study is outside the major interest of
the study, it is possible to see some of the frames of
reference within which Council members perceive the various
categorical roles by singling out certain items within In
strument B, and examining the responses to these items in
terms of what might be called their axial orientation.
In the professional community organization litera
ture, there is an attempt by Murray G. Ross to conceptual
ize the various approaches to community organization.®
ity was p > .30 < . .bO. Here again the test was not applied
to the Staff role (see footnote "b", Table b, p. 142.
O
Murray G. Ross, "Conceptual Problems in Community
Organization," Social Service Review, XXX (June, 19b6),
pp. 174-81. This idea is developed further in this same
author's Case Histories in Community Organization (New
York: Harper & Brothers, I^b8j7 pp. 3-13.
164
This has resulted in his setting up a classification sys
tem, referred to by him as a "scale," which includes the
following named approaches: "Exploitive Orientation,"
"Reform Orientation," "Planning Orientation," "Process
Orientation," and "Therapy Orientation." Ross specified
the characteristics of each of these orientations in ideal-
type fashion. Although he does not refer to them as such,
he makes it quite clear that he does not see them as mutu
ally exclusive, or as pure forms, as the following state
ment shows:
Each represents, conceptually, an approach separate
and distinct from the others; yet, in practice, they
may be merged in a single approach with planning and
action representing stages or elements of the process
orientation.9
The experience of the writer in the Council seemed
to suggest that the two axes at the extremes of Ross's
scale were not relevant to the Council situation, but that
the "Reform," "Planning," and "Process" orientations might
be. Further, it was hypothesized by the writer that there
might be a fourth axis around which the Council delegates
could become oriented, namely, the "Supervisory."
What characterizes these four conceptualized orien
tations— the three from Ross and the one suggested by the
writer? An attempt to set forth the key idea or the most
^Murray G. Ross, Case Histories in Community Organ
ization (New York: Harper & Brothers, ' 1958 J, p. 12.
165
salient feature of each follows:
1. Supervisory, views the Council as being a kind
of "over parent," to help insure adherence to established
social work practices and to the Community Chest and
Councils' policies, and to apply both formal and informal
sactions to help member agencies keep in line.
2. Reform, involves direct action and the sales
method of approach. One's mind is made up in advance as
to what the solution to the problem should be.
3. Planning, starts with a concern, and a commit
ment to the need for an exploration of the problem situa
tion. Committee work follows, which may result in recom
mendations and proposed actions, which are then accepted
or rejected by the larger body.
4. Process, is concerned with the developing of the
community's capacity to work on common problems. It mini
mizes the value of the specific plan or solution developed
at any given time, and maximizes the value of the growth or
maturation of the problem-solving and decision-making
skills and insights of the group members.
Two items thought to be characteristic of each of
the above axial orientations have been built into Instru
ment B as follows: Supervisory— items 2 and 12; Reform—
items 1 and 5; Planning— items 4 and 7; and Process— items
11 and 14. An analysis was made of the responses to these
items by the various role judges within the modal pattern
166
framework, which was utilized in an earlier portion of the
study (see pp. 91-95). One modification to this pattern
was made here, however. Although the relative item weights
as set by the degree of unanimity of perception within each
role judge grouping were utilized here, the actual agree
ment or disagreement with the particular items were taken
per se. (Previously, in the study, the "agree-disagree"
aspect was treated in terms of consensus.) For example,
if the respondent, when viewing his own role, agreed with a
statement which symbolized a given axial orientation, this
was interpreted as his having accepted this particular ap
proach to the Council. Likewise, if he perceived the in
cumbents in each of the other categorical roles as being
likely to agree with the same item, this was interpreted
as further favoring the given axial orientation. The net
score for each of the items was arrived at by subtracting
the "disagrees" (weighted values) from the "agrees" for all
three of the categorical roles.
It was hypothesized that "Planning" would rank first
and "Process" last, with the "Reform" and "Supervisory"
orientations sharing a somewhat equal intermediary position.
While no refined statistical tools were used to test
the statistical significance of the differences in the net
scores (since this phase of the study is not considered to
be crucial to the over-all study, but rather possibly sup
portive of other findings), it is reasonable to assume that
167
the considerable differences which are present might well
be significant. An examination of Table 12 shows a clus
tering around the "Planning" axis and a complete rejection
of the "Supervisory" axis among all of the role judge
groupings. Moreover, both the "Process" and "Reform" axes
showed greater scores than were anticipated.
TABLE 12
AXIAL ORIENTATION SCORES OF ROLE JUDGE GROUPINGS,
AS THEY PERCEIVE THEIR OWN
AND THE CATEGORICAL ROLES
Role
Axial Orientation3
Judge
Grouping
Supervisory Reform Planning Process
MSW
0 21 34 18
Non-MSW -4 4 39 2b
Laymen -5 13 28 13
Total -9 38 101 56
Plus scores show the extent to which the "agree"
exceed the "disagree" responses on the items which symbol
ize the particular axial orientation; minus scores, on the
other hand, mean that the "disagree" responses exceed the
"agree."
An inspection of Table 12 with regard to role judge
groupings reveals that MSW and Laymen appear almost equal
in their degree of acceptance of the "Reform" and "Process"
orientations; while the Non-MSW accept the "Process" axis
to a much greater extent than they do the "Reform" axis.
168
Some Likely Frames of Reference
Utilized as Role Perception Anchora'qes
~h ~as Revealed In Interviews
Interviews with laymen
As was stated earlier (pp. 121-26), the major pur
pose of including in the study design interviews with a
sample of Council delegates, was to provide a means of
checking the questionnaire data, and to search for possible
perceptual anchorages for the judgemental statements (re
garding the various categorical roles) which were checked
by the respondents. The study design called for thirteen
interviews with laymen, to be selected from both the long
term and the short-term participants, and from the non
respondents to the questionnaire as well as the respond
ents. The conclusions drawn from these interviews will be,
of necessity, subjective and selective, inasmuch as they
have not been subjected to any rigorous measurement criter
ia .
It should be pointed out that the individuals iden
tified in this study as "laymen" are not a homogeneous
category of persons in terms of their being exclusively
outside the social work profession. The layman classifica
tion represents a functional role in the Council only.
Some delegates function as such, but are actually earning
their living in the practice of social work. For example,
one lay delegate is on the Council by virtue of being a
169
member of the governing board of a non-case work agency, but
is actually employed by a county-level case work agency.
Consequently, he is functioning as a layman in the Council,
while his out-of-Council role is that of a social worker.
This kind of situation, of course, provides for potential
role confusion and perhaps role conflict, not only at the
role behavior level, but also at the role perception level.
Role of the layman.--As might be expected, there are
considerable variations in the responses of the laymen re
garding their own functions in the Council. Some appear to
be completely baffled by what is taking place; some express
a great desire to be of help, but do not know where or how
to take hold; some appear willing to function only nomin
ally, and to assume without question that the social work
ers know what they are doing, and will see that the Council
moves in the necessary direction; some appear quite sophis
ticated in their participation, and can not only see what
is going on but are also aware of the more subtle aspects
of the group process, i.e., the interpersonal conflicts,
the existent unilateral agreements, the professional biases,
and the agency-vested interests.
Let us look, first of all, at some of the responses
of laymen who appear to feel reasonably secure in their
Council roles, who give some evidence of comprehending the
value of the work of the Council, and who seem to have some
170
understanding of the group processes in operation. This
segment of the interview sample was composed mostly of per
sons in the field of business. (One exception was a per
son serving as a lay delegate, who is employed by another
agency. ) These laymen appear to have a rather clear-cut
idea of why they are on the Council, although their reasons
differ from one another. Some cases will be singled out as
illustrations of this generalization.
There is, for example, one person whose status in
business circles is high, who has considerable respect for
the social worker and a great appreciation for his skill
and professional training, but who expresses a concern for
an "occupational disease" which the social worker presum
ably has, namely, "the tendency to want to extend services
to everyone." This layman is concerned about "doing too
much for people," and hence removing from them the sense of
responsibility for solving their own problems. He certain
ly favors sending social workers to conventions and work
shops, and even helping to pay their expenses for such, but
complains that they often come back so enthused about a new
service that they want to adopt it immediately in their own
agency. This layman, therefore, apparently sees his own
role in the Council as being one of exercising a check upon
this propensity, which could possibly be interpreted as
merely a rationalization for not wanting to spend money for
social services. However, such an explanation did not seem
171
to quite fit this respondent. More likely, his Council
role perceptions are anchored in his total view of society
and in some of the dangers with which he may believe it is
confronted. He does not want his present way of life to
disappear, and this "occupational disease" which, in his
judgment, is characteristic of social workers, constitutes
a possible threat to this. He, therefore, sees the Lay
role in the Council as a safeguard against social workers'
inclinations to expand the scope of their work. It is
interesting, however, to note that this layman does not
have a high role involvement score as measured by the in
volvement index. This might be explained by his statement,
"The social worker will naturally act in a more responsible
manner if he knows someone is present to check on his rec
ommendations." Attendance at meetings, therefore, seems
to constitute, for him, his major function in the Council.
There is a further clue to the frame of reference
within which this person apparently perceives his role. A
business man must, in his own judgment, be concerned about
community welfare enterprises, such as the Council repre
sents. In his particular business, a division of labor has
been worked out in this regard. "My boss and I do most of
it in our establishment," he states. This layman's insight
into the subtle implications which volunteer civic work has
for the business-man role is incisive. "You can't do civic
work at the expense of your job," he insists. "You have to
172
be able to do both well." He further emphasized the impor
tance of handling a volunteer assignment well by stating,
"It soon gets around (in business circles) if you don't do
a good job." Here is a strong inference that his status in
the business-man role is very closely related to his per
ception of the layman role in the Council. On the other
hand, this same man expressed considerable bewilderment
when it came to discussing his aid in making policy decis
ions within the Council. He seems to feel personally inad
equate, and indicated that he believed most of the decisions
rest within the social workers' areas of competence and are
not, in the main, within the layman's scope of knowledge or
experie nee.
Another business man's point of view will be de
scribed somewhat in detail, since it seems to be close to
the idealized norms and expectations expressed in the pro
fessional community organization literature concerning the
layman role. This layman sees the business man as posses
sing an approach to problem-solving which, in his judgement,
should be brought to bear upon the community welfare plan
ning process. According to this layman, this approach in
volves, in his words, "a directness," "an objectivity,"
"a respect for hard facts," and an awareness of the "over
used and under-explored financial and human resources"
within the community. There appears to be nothing of the
"watch-dog" frame of reference in his perception of the
173
layman's relation to the social worker. "I have a bent
toward spending (money) for what is necessary in the commu
nity," he declared, and he believes that the only way to
know what is really "necessary" is for the professional
worker and the layman to arrive at this jointly. He says
he sees his Council role as composed of three interrelated
sets of functions: (1) to engage in this decision-making
process by bringing the business man's approach to problem
solving into the welfare-planning process; (2) to help make
business aware of the human values involved in welfare
services, so that business men will be willing to help de
fray the necessary costs (he sees himself as a kind of am
bassador from the welfare-conscious community to the busi
ness community); and (3) a function which involves his role
on the agency board which he is representing in the Coun
cil. Concerning this latter function, he said, "Serving
on the Council enables the board member to see the agency
in the light of the total community."
These three factors are precisely what the designers
of welfare councils indicate that they hope to accomplish
as they involve the layman in such work. It is interest
ing to note that the layman just described does not, as was
also true in the previous case, have a high role involve
ment score as measured by the involvement index, in spite
of being active on the board of the agency he represents on
the Council. Further, it should be noted that his contact
with a professional social worker is very close and fre
quent, which may help to account somewhat for his compre
hensive view of the idealized layman's role.
Some laymen feel that the layman’s chief contribu
tion is that of "asking questions." One man, who sees
this as a valuable function since it forces the clarifica
tion of issues through the discussion process, further ex
pressed the view that social workers need to be forced to
think through their recommendations and procedures in terms
that laymen can understand.
Another layman views the Council as a kind of "lis
tening post" for ideas to be carried back to the organiza
tion he represents. (His organization adopts certain
projects each year, and then seeks to raise money for their
support. )
Still another viewpoint was expressed by a layman
who supports the Council strongly on the ideological level,
but admits to being very confused by its actual workings.
He likewise is a business man, and is not in sympathy with
those in business who pay little attention to welfare ac
tivities in the community. "I am part of this community,"
he says, "and I take profits out of it." He stated emphat
ically that he, therefore, had a responsibility to "put
something back into it" (the community). "I'm concerned
about what happens to this town," he further commented, and
then went on to relate how, as a young man starting out in
175
business, he was advised by his banker to become identified
with some recognized civic activity. To his banker, such
activities seemed to symbolize stability, "a stake in the
community," and perhaps even a cue as to who constitutes a
good economic risk in terms of loans. This layman spoke
disdainfully of the businessman who does not take the
trouble to become concerned with community welfare activi
ties. He gave the impression that he felt that being a
good businessman is equated, in part, with being involved
in civic enterprises, and that anything less than this is
being short-sighted in matters of business. This layman
expressed difficulty, however, in finding a place to take
his part in the Council, although this does not appear to
be the case in the particular agency board on which he
serves. He pointed with considerable pride to his being
able to save that agency money by giving advice and coun
sel on matters of plant construction and maintenance.
This is the area in which he apparently feels he can make
his greatest contribution. It is possible that the agency
which he represents has named him as its delegate to the
Council with the hope of enlarging his view of welfare
services in the community. With such high motivation to
serve, however, and yet no place to "take hold," he is ex
periencing considerable frustration.
Another layman was not only baffled by the opera
tions of the Council, but did not seem to have any notion
176
whatsoever of its function, or even how he came to be ap
pointed to the Council. To illustrate the extent of his
confusion, this individual was perplexed by the term
"case," as it applies to case work services, and seemed
able to think only in terms of the word as it is used in
the field of law. Since this man, however, was among the
non-respondents to the questionnaire, his confusion is not
represented in the questionnaire findings.
Another segment of the lay grouping somewhat hostile
to the Council thoroughly recognizes and accepts the worth
iness of its objectives. This hostility seems to be an
chored in the perception of the way the Council is conduc
ted, and the manner in which its agenda is determined. One
expression of this was, "I simply come for lunch and sit;
that's all. My presence neither enhances nor hinders the
agency 1 represent, or the way decisions are made in the
Council. The work is all cut and dried; the social work
ers dominate the meeting; and the whole thing is totally
outside the layman's hands." This person further stated
that if a layman did make his voice felt in the decision
making process, the professional workers would construe it
in a manner that represented their own desires and wishes.
This man, who likewas was a non-respondent to the question
naire, attends Council meetings infrequently.
Not all of the hostile interviewees, however, were
as detached from the work of the Council as this last one.
177
One layman (long-term) with a high role involvement score,
insisted that his long tenure had not been rewarded with
important committee assignments, and that the reason for
this was his frequently verbalized dissent of the decisions
made in the Council. "The Council is simply window dress
ing," he insisted. "Regardless of what goes through the
Council or into it, whatever comes out is only what _________
(name of a person) permits." This interviewee’s self per
ceptions are that of "a rebel" and "a free spirit." He
decrys the manner in which individuality is being suppressed
in our society generally, and the manner in which the power
structure prohibits freedom of expression and independence
of thought. According to him, adjustment is the key con
cept in our society. "Schools teach it, labor unions de
mand it, and the political party not only demands it but
rewards it, so what more can you expect." The Council, for
him, is no more than a "rubber stamp." This layman, how
ever, appeared to be thoroughly conversant with the vari
ous projects that had moved through the Council, and very
much aware of many of the implications of the recommenda
tions that it had made. In his case, the major frame of
reference for Council role perceptions appeared to be fur
nished by the broader societal orientation, and the sense
of "personal mission" which seemed to characterize his self
image.
178
Some concluding generalizations about the layman
role.— Although subjective and selective in nature, the
above descriptions suggest some possible, or even likely,
frames of reference within which the laymen may be per
ceiving the various categorical roles in the Council.
There is no evidence, in the interview data, that
the lay delegates perceive the Council roles within a com
mon frame of reference. Only a limited number of them have
been subjected to any acclimitization processes preparatory
to their taking on the Layman role. Moreover, the Council
structure does not appear to provide its lay members with a
uniform set of role expectations, as length of tenure in
the Council seems to make very little difference in this
regard. Whenever clarity of role prevails, it would ap
pear that it is due more to processes and forces outside
of the Council structure, and related to inter-personal
contacts which are somewhat incidental to the performance
of the lay role within the Council.
From the impressions gained from the interviews with
laymen, four conclusions are suggested:
1. There is perhaps more anchorage of the layman*s
role in his own professional role outside of the Council
than the questionnaire data seems to indicate. (It is
possible that the questionnaire was not sufficiently sen
sitive at this point.)
2. There appears to be some support for the notion
179
that the Lay role in the Council is considered to be of
value to the businessman role.
3. There seems to be acceptance of the general
format of the Council and of its work, but some feeling
that the layman role is void of specificity of expecta
tions .
4. There seems to be considerable awareness of the
implications of what goes on in the Council and its rela
tionship to the larger values and norms in the total soci
ety (referred to in this study as "Broader Societal").
Interviews with social workers
As social workers were encouraged to talk about
their own roles in the Council, one active frame of refer
ence became evident. From the interviews, it would appear
that the social worker *s Council role is strongly anchored
to his agency role, and that the expectations surrounding
the agency role constitute for him a major frame of refer
ence within which he perceives his own Council role. This
does not mean that the two roles are fused or that one is
simply an extension of the other. The two roles appear to
be recognized by social workers as separate entities, yet
the norms and expectations surrounding the delegaters
Council role appear to be perceived constantly in the light
of the norms and expectations of his agency role. To
state the matter in another way, there appeared to be no
180
evidence that the social worker transcends the more limited
view of his own agency and takes on a broader community
point of view as he engages in the community welfare plan
ning process within the Council framework.-^ These gener
alizations grow out of a number of different kinds of rela
ted ideas expressed by the social workers, which will be
discussed briefly.
Most of the participating social workers see their
Council role as making them more efficient in the perform
ance of their social agency functions. This sometimes was
expressed in terms of their being more informed about com
munity resources which could be utilized in client needs
and, at other times, was expressed in terms of seeing the
functions of their own agencies in the light of related
agencies. "Gaining information about other agencies and
their programs," was often stated as one of the advantages
of serving on the Council. Moreover, it was frequently
stated that the delegates expect to be, and are, called on
to report on the work of the Council in agency staff meet
ings.
10In our culture, the concept "delegate" usually im
plies "representation" which, in turn, implies two inter
related norms: (1) that the delegate look out for the inter
ests of the parent group; and (2) that the delegate be per
mitted the freedom, on certain occasions, to act in the in
terest of the larger community if his judgment so dictates,
even though this may call for the temporary setting aside of
the specific interests of the parent group. Usually this
latter norm comes to be expressed at the idealized level.
181
Another aspect of the interrelationship of agency
and Council roles demonstrates, in somewhat more subtle
terms, the anchorage of the Council role to the agency
role. Social workers are highly conscious of their agency
role, as they function in the Council, seldom seeming to be
very far removed from it, and there is an evident caution
surrounding the statements they make in Council meetings.
One long-tenure delegate said, "I seldom can speak as an
individual in the Council. I am known as _______ (his
agency position title). I must always be aware of how my
statements will be taken in terms of this (latter) role.
Sometimes this requires that I speak up, and sometimes it
means I must say nothing though I may feel strongly on the
matter personally."
Another long-tenure social worker participant made
a similar statement, "I speak up only on matters related
to my job. This is all," He continued with a most inter
esting and suggestive observation. "I know what to expect
of a person in the Council when I know what agency they
represent." Representatives of state and federal agencies
appear to be especially aware of the legal and policy
limitations placed upon their pronouncements and behavior
as they participate in Council activities.
For some of the delegates who occupy policy-forming
positions in their respective agencies, the Council role is
seen as a way to find out, as one person expressed it,
182
"what the community is willing to accept and go along
with." The Council is thus perceived as a kind of sounding
board fox policy making and as being helpful in the per
formance of the agency role.
All of the social workers seemed to be aware of the
need for representing their own agencies well, and of their
obligation to speak in Council or committee meetings when
matters affecting their own agencies were being consider
ed. Most social workers expressed a feeling of obligation
to participate in the discussions whenever their agency
experience provided them with pertinent knowledge. How
ever, this knowledge or experience was always spoken of in
terms of the agency role and not in terms of the citizen
role of the delegate.
One set of observations which grew out of the inter
views with the non-MSW role judges, covered an often ex
pressed concern about such things as how the social worker
is being viewed by the layman, whether the Council is suf
ficiently broad in its coverage, and the necessity of so
cial workers to transcend the boundaries of specialized
services. Questions were raised by them as to whether it
would be possible for the Council to do a better job of
ferreting out the unmet needs in the community. There ap
peared to be somewhat more concern among the non-MSW than
among the MSW role judges about the profession of social
work as such, and its relation to the total community and
183
the other professions. Some of the non-MSW delegates
showed signs of having gone through the period when social
work was less well accepted among the professions than it
is today. This manifested itself in the tendency to iden
tify with the struggle to become a profession, and in the
satisfactions resulting from hard-earned recognition. Some
of these non-MSW persons seem to anchor their Council role
perceptions in the obligations that adhere to members of
a profession as much as in their agency role.
Social workers1 views of laymen in the Council.--
None of the social workers interviewed questioned the value
of having the layman participate in the activities of the
Council. There seemed to be a strong acceptance on the
part of the social workers that "the layman must be brought
along" in the community organization process. Including
the layman in the community welfare planning process, ap
pears to be well-sanctioned by the social worker's profes
sional training, and the validity of the idea seems to be
supported by the social worker's experience in agency work.
The following excerpts, taken from the interview material,
illustrate some of the views expressed with regard to the
necessity for including laymen in the planning process:
"professionals can decide, but they can't make it happen
(alone);" "the layman has a practical outlook;" "the layman
provides a much needed link to the community;" "he helps
184
social workers keep of cloud eight;1 1 "the layman can sway
a segment of the community;" "the reality of paying for
services must be faced;" "social workers need some re
straints on over-programing;" "social workers tend to de
liberate too long, whereas a businessman helps speed up
decision-making, and asks for action." One of the Non-MSW
role judges raised the question of whether the profession
of social work might not have "oversold" social work to the
layman, beyond its ability to "deliver," and whether this
might not be giving the layman some concern as he views
social work at close range. Another non-MSW delegate ex
pressed great confidence in the lay view as such, and de
fended the layman's "right to ask questions about social
work practice." Another person from the same group ex
pressed doubt as to whether the Council had really accepted
the idea of including laymen, together with social workers,
in the community organization process.
The fact that the layman is viewed by most social
workers as essential to the community organization process
seems to stem, at least partly, from the training which
social workers have received, i.e., they seem to accept
this as part of the dogma of their profession. Occasion
ally, there was an expression such as, "The layman must be
terribly confused by all this business." Yet there seemed
to be no concern about finding out whether the layman views
his own role as sufficiently well structured to make his
185
participation in the Council a really satisfying experi
ence. It did not appear that the layman was viewed by most
social workers as having intrinsic value in the decision
making process in the Council. It seemed, rather, that the
major frame of reference for his being viewed as essential
to the welfare planning process has two other major foci,
namely: (l) the layman in the Council is perceived as help
ing to guarantee the necessary support for welfare services
in the community; and (2) he is perceived as helping social
workers sense the community tolerance limits for policy
formation and proposed extensions of services. Apparently,
the social workers in this setting have, in the main, ac
cepted the idea that including laymen in the Council is one
of the facts which must be taken for granted. They do not
appear to be aware of the unstructured nature of the lay
mans role which this study reveals.
Summary
Instead of summarizing the findings of each tested
hypothesis in the order in which it was presented in this
chapter, some of the findings will be rearranged for the
sake of brevity and clarity. In short, the questionnaire
findings, as they have been brought to bear upon the re
search hypotheses, are as follows:
1. The time element (length of Council tenure) did
not prove to be a statistically significant
variable with regard to:
a. The degree of role consensus among Council
delegates, irrespective of role judge
groupings.
b. The degree of role consensus within each
role judge grouping.
c. The degree of stated role clarity of Coun
cil delegates, irrespective of role
judge groupings.
d. The degree of stated role clarity within
each role judge grouping.
e. The extent to which intra-Council factors
become a source for role perception an
chorages .
f. The extent of role involvement of Council
participants, irrespective of role judge
groupings.
Role consensus proved to be highest among the
MSW grouping, somewhat less among the non-MSW
grouping, and least among the laymen.
There is a consistency among role judges con
cerning the "goodness of role perception,"
i.e., good role perceivers of their role of
incumbency tend to be good role perceivers of
the other Council roles, and poor role per
ceivers of their role of incumbency tend to be
poor role perceivers of the other roles.
There does not appear to be a significant dif
ference, in the degree of stated role clarity,
between social worker and lay role judges.
Role involvement is greatest among the MSW role
judge grouping and least among the laymen,
with non-MSW grouping approximating the lay
pattern.
Social workers tend to anchor their perceptions
of the Social Worker role, the Staff role, and
their own (personal) role in the professional
orientation syndrome; whereas laymen tend to
anchor their perceptions of the Lay role in
the broader societal orientation, and their
own (personal) role in the intra-Council ori-
entat ion.
The MSW role judges are more professional-train
ing oriented than are the non-MSW role judges,
when perceiving the Social Worker and Staff
categorical roles, but this is not the case
when they are perceiving their own (personal)
role.
The Staff role is perceived by all role judges
as being the most highly structured role in
the Council, the Social Worker role as being
somewhat less well structured, and the Lay
188
role as being the least well structured role.
9. There does not appear to be a statistically sig
nificant relationship between the degree of
role involvement of the Council delegates in
the Council enterprise, and the degree of the
f ollowing:
a. Stated role clarity.
b. Role consensus.
c. Stated Council role satisfaction.
d. Stated Council role interest.
10. There does not appear to be a statistically sig
nificant relationship between role consensus
and stated role clarity, irrespective of which
categorical role is being perceived.
11. An interpretation of certain questionnaire items
strongly suggests that Council delegates view
the Council functions in terms of the "Plan
ning" axis, and totally reject the "Reform"
axis. The MSW and lay judge groupings are
about equal in their acceptance of the "Re
form" and "Process" axes, while the non-MSW
accept the "Process" axis to a much greater
extent than they do the "Reform" axis.
With reference to the interviews, one thread appears
to run through the comments made by the laymen, which is an
expressed concern about the implications of social work
189
practice and philosophy for the total society and for the
immediate community, as perceived by the layman. This con
cern seems to constitute the major frame of reference with
in which the layman perceives the various roles and opera
tions of the Council. In addition, the Council structure
does not appear to be furnishing the lay delegates with a
common set of role expectations for the various Council
roles, and there appears to be some support for the notion
that the Lay role in the Council is considered to be of
value to the businessman role.
It would appear that the social worker's Council
role is strongly anchored to his agency role and that the
expectations which surround the agency role constitute for
him a major frame of reference within which he perceives
his own Council role.
CHAPTER VI
CONCLUSIONS
Some Interpretations and Implications
of the Study
Certainly the findings of this study further justify
the question raised in Chapter II concerning the assumed
"postulate of consensus" which characterized some of the
early writings on social role. This study points conclu
sively to the fact that, when explanations for deviant or
conforming behavior are being sought, deviant behavior may
actually be interpreted as differing perceptions of role
expectations. This suggests, of course, that perceptions
should be taken into account— together with motivations,
attitudes, and other personality factors— as intervening
variables which influence role behavior.
Further, the study reinforces the belief that the
process of perception, which is linked directly to the or
ganism, is structurally influenced (speaking social-psy-
chologically) not only by the group in which the behavior
takes place, but also by other groups in which the indivi
dual has membership and/or in which he aspires to be ac
cepted. (This is, in essence, the concept of "reference
group.") The concept of reference group, as a factor
190
191
in role perceptions, may help explain one of the findings
of this study, which was quite contrary to the hypothesis
posited at the outset. It will be recalled that one hy
pothesis suggested that long-time participants would show
greater role consensus than short-term participants, and
this was not supported, whereas ordinarily this might be
assumed to be the case.-*- There appear to be some factors
operating in the Council which neutralize the factor of
time to such an extent that it alone is not as an important
a variable as it was hypothesized to be. What explanations
may be offered?
It is possible that multiple group membership, as
combined with the time variable, complicates the role per
ception process among Council delegates. In fact, Gross,
Mason, and McEachern, in their study of the superintendency
and board member roles, appear to support this notion. Com
menting on their own findings, these authors state:
When X and Y first interact they may hold expecta
tions for each other based on their perceptions of each
other's occupancy of a single position, for example,
that of assemblyman in the workroom. As X and Y inter
act over a time, other positions they occupy may enter
into their perceptions of each other. X learns that Y
is married, a father of three children, and a Catholic;
and Y learns that X is a bachelor and does not belong
to any church. As these further identifications become
known they may influence the type and nature of inter
action between X and Y and the evaluative standards
■^Expressed in General Hypothesis 3, and in Research
Hypotheses 1, 4a, and 5a.
192
they apply to each other's behavior.^
In interpreting the above findings, Stogdill makes
the following statement:
Gross, Mason, and McEachern found that the degree
of consensus within each sample relative to the attrib
utes required for potential occupants of a position was
significantly greater than that relative to the per
formances required after occupancy of the position. .
. . These results appear to lend direct support to the
hypothesis that those expectations which define posi
tions exhibit a higher degree of mutual confirmation
and stability than do those that define roles.3
In the light of those comments, and taking into ac
count the findings of this study, two factors have been
overlooked which might be important in further role stud
ies: (1) the effect of social positions and status ratings
on the degree of role consensus; and (2) the effect of time
as it increases the potential for multiple reference group
influences upon the perception of roles being studied.
Gross, Mason, and McEachern also found that consen
sus between the superintendent and the board members, in
defining their roles, varied with the ratings of the board
member by the superintendent, i.e., the higher the super
intendent's rating, the higher the consensus as to the role
o
Neal Gross, Ward S. Mason, and Alexander W.
McEachern, Explorations in Role Analysis: Studies of the
School Super intendency Role (New York: John WiTey arid Sons
Tnc.ri'^aTT p. STH-’ Sfr:-----
o
Ralph M, Stogdill, Individual Behavior and Group
Achievement (New York: Oxford Universlty Press, 19!o^J, p.
T5T.
193
expectations of the board member.^ Again, this suggests
that perhaps the factor of social position should be given
a larger place in the study of role perceptions than the
present study provided. It may be that a high status per
son is allowed a certain freedom to perceive individually,
rather than in terms of the usual role norms; or, on the
other hand, the high status factor may merely serve to
identify a person who has a greater variety of reference
groups upon which to draw for his perceptual frames of
reference and consequently expresses a greater range of per
ceptual judgments. With regard to the present study, then,
it is possible that the long-term participants (3+) did not
show greater consensus than the short-term ones (2-) be
cause: (1) their being in the Council longer might mean
that they are high status persons, and (2) their longer
time in the Council may have provided the necessary ele
ments for them to feel free to individualize their percep
tions of the various roles and, consequently, to demon
strate less consensus than the 2- participants.
Gross, Mason, and McEachern also discovered that
there was more consensus among board members whose motives
for seeking elections were similar, than among those with
dissimilar motives.^ In view of this, the factor of moti-
^Gross, Mason, and McEachern, . ojd. cit. , p. 217.
5Ibid«. p. 187.
194
vation, which was not controlled in the present study,
might be an additional factor in the decreased amount of
consensus among the 3+ participants. Long tenure in the
Council, might suggest a greater potential range of motiva
tional factors.
One of the findings, reinforced in many different
ways throughout the study, is that it is the role being
perceived, rather than the type of role judge doing the
perceiving, which appears to be significant. Roles that
are well-structured are perceived by all as being so, and
roles that are poorly structured are perceived by all as
such. This lends support to General Hypothesis 1, which
held that the roles within the Council would be perceived
differentially. The Staff role, for example, is judged by
all as being the most clearly structured of the three cate
gorical roles, and it is so judged by all regardless of
other variables, such as length of tenure, role involvement,
etc. The Lay role, on the other hand, is perceived by all
as being conspicuously lacking in structure. In the case
of the Lay role, however, it does tend to be perceived by
the layman as being more clearly structured as the length
of time in the Council increases, and as the other varia
bles increase in magnitude.
If Newcomb is correct in his conclusion that role
expectations perform an important function in the inter
communication process in the group, then the lack of
195
specificity in the structuring of the Lay role would cer
tainly appear to be in need of attention by those respon
sible for the administration and leadership of the Coun
cil. ^ In the event that Council leadership develops an or
ganized concern about the indefiniteness of the Lay role,
and that the group itself decides to do something about it,
there is evidence in this study that an important element
is already present which should help insure positive re
sults. Laymen show considerable consensus in perceiving
the role of the other, and role theory suggests that the
perception of the role of the self is dependent upon the
perceptions of the role of the other (see Chapter II). In
view of the fact that consensus of the role of the other
already exists to a marked degree, it should not be too
difficult to structure the Lay role more precisely.
It may be that the need for further study is sug
gested by the fact that the Staff role is perceived as
being so highly structured. True, there is only one staff
person in the Council, and the way the Council functions
makes this role a central, though not necessarily a leader
ship, one because most Council activities are carried on
through and around this central role.
A
Theodore M. Newcomb, Social Psychology (New York:
Dryden Press, 1950), pp. 293-9T. See also Kingsley Davis,
Human Society (New York: The Macmillan Company, 1949), p.
■ st:
196
It might be well to ask, at this point, what ingredi
ents of the perception process are present which enable all
Council members to "categorize this particular role," as
Bruner would characterize it (see Chapter II). What en
ables the layman to round out his perception of the Staff
role, or what model does he use to "size up the situation,"
so that the end result is a fairly uniform psychological
product? Does he use the "managerial" model, the "strong
leader" model, a kind of generalized "executive secretary"
model, or is there something with the Council itself which
furnishes him with a category into which to fit the behav
ior of the staff member and thus to perceive it as others
do?
Another important finding is the one indicating that
role incumbents tend to differentiate between their own
(personal) role and the categorical role in which they have
their incumbency. This is particularly true with refer
ence to the laymen. Social workers tend to see both their
own and the categorical role of which they are a part "more
clear than unclear." Laymen, by contrast, see their own
role as role as being "more clear than unclear," but see
their categorical role as "more unclear than clear." This
could suggest that the layman may be thinking, "Of course,
I know what I'm doing, but those other laymen don't." Or,
he might be thinking the reverse of this, such as, "I don't
know what it is all about, but, of course, the other laymen
197
do."
On the other hand, this differentiation between own
and categorical roles could suggest a lack of identifica
tion with the role to the point of holding the "other"
responsibile for actions from which the "self" is being
exempt. If this is true, role confusion might be a likely
result.
Still another situation revealed by the findings
which contributes to possible role conflict, is the fact
that the non-MSW grouping, being the most varied in its
role perceptions, is often unlike either the MSW or the lay
group. Moreover, when it does approximate the perceptions
of one or the other role judge groupings, it usually approx
imates the lay perceptions rather than those of the MSW.
They must still be classed as social workers, even though
they may perceive, at times, more like laymen, because they
must function as social workers. This suggests some poten
tial role conflict, but it also suggests the possibility of
the balance of social power residing within this non-MSW
grouping of participants. This is an implication with
which the present study is not concerned.
This study seems to justify the use of the role-cue
statement as a quick way of getting at role perceptions.
Respondents seemed to have no difficulty attaching most of
the role-cue statements used in the questionnaire to one of
the specific Council roles. There was sufficient consensus
198
among most of the role judges to indicate that this was not
a random or chance phenomenon, but rather that the judg
ments were patterned. Moreover, the patterning appeared to
be consistent, in that the three instruments in the Role
Consensus Battery tended to reinforce each other.
Some Suggestions for Further Study
The findings of this study suggest several kinds of
follow-up studies which may be profitably made. Some of
these will be mentioned briefly.
First, there is a growing literature, within the
fields of sociology and social psychology, dealing with the
institutionalized aspects of the professions and occupa
tions. This work, along with the study of bureaucratic
structures and large scale organizations, could well be
duplicated in the field of social work and with reference
to social agency structures. Social work provides an ex
cellent opportunity for this kind of study, as it is one of
the newer professions, and is presently working toward be
coming a licensed one. It is a profession in the "process
of becoming," and the dynamics of this process could throw
considerable light on the development of the profession as
such.
Further, the existing body of role theory would seem
to provide ample support for numerous explorations into some
of the different facets of role expectations and percep
199
tions in the community organization field, as the field is
conceptualized by the social worker. The following sugges
tions are submitted as examples:
1. A role content study of the council roles, and
the extent to which this content overlaps, or is in con
flict with the agency, professional, and community roles.
Such a study would help to establish the validity of the
present study still further.
2. A before-after study of role perceptions, where
an attempt would be made to clarify the role expectations
of the various council roles (particularly the lay role),
and then measure the differences in role perceptions after
this attempted clarification had been made.
3. A study of role perceptions as they are related
to personality factors, as measured by some appropriate per
sonality inventory test.
4. A role study in the general field of community
organization, where the unique elements of the local situa
tion would be minimized, and the elements common to the en
tire field generally could be appraised. Such a study
should help in the further professionalization of the field
of community organization, and should be helpful in the
training of community organizers.
5. A role study of the lay-professional relation
ship, which would go beyond the boundaries of community
organization, and deal with these two roles in different
200
settings, such as in the fields of health, city planning,
educational planning, etc.
6. A role perception study where the role judge
groupings would be based upon commonality of values, philos
ophy, and attitudes. The latter variables would be treated
as independent variables.
7. A role study in community organization, where
the similarities and differences between ideal role expec
tations (as reflected in the professional literature in the
field), and real expectations (the actual behavior of the
person in the role, as this is reflected in the day-to-day
operations in the work situation) could be ascertained, and
the awareness of the participants to these similarities and
differences could be measured.
BI BL IOGRAPHY
BIBLIOGRAPHY
Barry, Mildred C. "Current Concepts in Community Organiza
tion," Annual Forum of National Conference of Social
Work, Group Work and Community Organization. New York:
Co1umbia University Press, 1956.
Blackwell, Gordon. "A Theoretical Framework for Sociologi-
, 1 cal Research in Community Organization," Social
Forces. XXX (October, 1954), 57-64.
Gross, Neal, Mason, Ward S., and McEachern, Alexander W.
Explorations in Role Analysis: Studies of the School
Superintendencv Role. New York: John Wiley & Sons, Inc.,
1958.
Haire, Mason. "Role Perceptions in Labor-Management Rela
tions: An Experimental Approach," Industrial and Labor
Relations, VIII (January, 1955), 204-'216’ . ' " '
Levinson, Daniel J. "Role, Personality, and Social Struc
ture in the Organizational Setting," Journal of Abnor
mal and Social Psychology. LVIII (March, 1959), TtcT™"
Merton, Robert K., Reader, George G., and Kendall, Patricia
L. The Student Physician: Introductory Studies in the
Sociology of Medical Education. (jambrldge: Harvard
l) n ive r sit y l>r e s s ~ ~ t 9^7.
Motz, Annabelle B. "Role Conception Inventory: A Tool for
Research in Social Psychology," American Sociological
Review. XVII (August, 1952), 465-471.
Neiman, Lionel, and Hughes, J. W. "The Problem of the Con
cept of Role - a Resurvey of the Literature," Social
Forces. XXX (December, 1951), 141-149.
Newcomb, Theodore M. Social Psychology. New York: Dryden
Press, 1950.
Reissman, Leonard, and Roher, John H. (ed.). Change and
Dilemma in the Nursing Profession; Studies of Nursing
Services in a Large General Hospital. ' New York: Put
nam, 1957.
Ross, Murray G. Case Histories in Community Organization.
New York: Harper & Brothers, ~ ‘ ^
202
203
Sarbin, Theodore R. "Role Theory," Handbook in Social
Psychology, Vol. I. Edited by Gardner ^indzay. Cam
bridge: Addison Wesley Publishing Co., 1954.
Sargent, S. Stansfield, and Williamson, Robert C. Social
Psychology. New York: The Ronald Press Company, Jl'^58.
Sherif, Muzafer. "The Concept of Reference Groups in Human
Relations," Group Relations at the Crossroads. Edited
by Muzafer She'rif and 6. WiTsorr! New York: Harper
& Brothers, 1953.
Stogdill, Ralph M. Individual Behavior and Group Achieve
ment . New York: 6x£ord University Press, l9o9.
Stumpf, Jack Eugene A. "The Role of a Community Organiza
tion Worker as Revealed by a Process Record of a Wel
fare Committee." Unpublished Master’s thesis, School
of Social Work, University of Southern California,
1949.
Thomas, Edwin J. "Role Conceptions and Organization Size,"
American Sociological Review, XXIV (February, 1959),
30-3'7.
Towner, Stanton B. "Role Student Nurses Desire or Expect
to Perform and Roles Achieved by Graduate Nurses from
the Same Selected School of Nursing." Unpublished
Doctor's dissertation, Department of Sociology, Univer
sity of Southern California, 1957.
Turner, Ralph H. "Role Taking, and Role Standpoint, and
Reference-Group Behavior," American Journal of Sociol
ogy. LXI (January, 1956), 316>-3^>8.
Wilkening, Eugene A. "Consensus in Role Definition of
County Extension Agents Between the Agent and Local
Sponsoring Committee Members," Rural Sociology, XXIII
(June, 1958), 184-197.
Statistical References
Guilford, J. P. Fundamental Statistics in Psychology and
Education. New Yorlc: McGraw-Hill Boole Company, Inc. »
T5 5 T.------
Hagood, Margaret Jarman, and Price, Daniel 0. Statistics
for Sociologists. New York: Henry Holt and~Company,
204
Jahoda, Marie, Deutsch, Morton, and Cook, Stuart W. Re
search Methods in Social Relations. New York: The
Dryden Press, 1951.
Siegel, Sidney. Nonparametric Statistics for the Behavior-
al Sciences. New York: McGraw-Hill Book Company, Inc.,
— —
Walker, Helen M., and Lev, Joseph. Statistical Inference.
New York: Henry Holt and Company" l£t>3.
A P P E N D I X E S
APPENDIX A
THE QUESTIONNAIRE AND OOVER LETTER
SENT WITH THE QUESTIONNAIRE
APPENDIX A
THE QUESTIONNAIRE
Code No.
A. PERSONAL DATA SHEET
1. Agency or Organization You Are Representing in the
Council:
Its name is ______ / What is
your present connection with it?
/
How long have you been thus associated with it?
/
Are you aware of any limitations or qualifications
which it places on your council participation? Yes^
No / If ves. please specify
•
Your occupation or profession:
Firm or aqency workinq for now
/
Title of present work
/
Housewife / Retired / Membership in
or occupational orqanizations
professional
Membership in business organizations
3. Education and training:
Highest academic degree or attainment /
207
Year completed / Any professional or occupation
al training not reflected in the degree title?
(Give amount & kind) __________________________
Council Participation:
Number of years on the council / Offices held
/ Estimate the number of com
mittees on which you have served . . (if more than
ten simply check here ____ /
Community:
List (other) welfare agency Boards of which you are a
member __ ___ __
Of all the organizations to which you belong name as
many as you can (up to five) in which you are really
proud to have membership
How would you classify the neighborhood in which you
have your residence? Working class / Lower Middle
Class / Upper Middle Class / "Upper Class /
B. ODUNCIL FUNCTIONS AND POLICIES
Here we have some hypothetical statements about council
functions and policies. We would like you to give two
types of reactions to these statements: l) what you think
would be the general opinion of the social workers on the
council, the laymen on the council and council staffj and
(2) your own personal opinion on each statement.
Directions: Please circle A (agree) or D (disagree)
in each column after each statement. Example: if you think
social workers on the council, and in the main, would agree
with the statement then circle A in the column labeled
"Social Worker." If you think laymen on the council, on
the whole, would disagree with the statement circle D in
the "layman" column. A or D circled in the "Staff" column
would reflect your estimate of how staff would likely view
the statement. However, A or D circled in the "My Own"
column would reflect your personal reaction to the state
ment. Be sure to circle one sy mbo1 1n each of the four
columns for each listed item.
Worker Layman s^aff My Own
1. The council should seek out
sources of demoralization and A D A D A D A D
social breakdown in the commun
ity and expose them.
2. The council should review
periodically the functions of its
member agencies and see if appro- A D A D A D A D
priate standards of case work
practice are being maintained.
3. If a given member agency
feels it has occasion to complain
about the practices of a related A D A D A D A D
agency its case should be re
viewed by the council.
4. Most council action should
be contained within this frame
work: Making studies of carefully
selected problems of member-
agency concern, building recom- A D A D A D A D
mendations on such studies, pass
ing these recommendations along
thru existing channels to the ap
propriate agencies, groups, or
persons.
210
Social Layman Staff My Own
Worker
5. Any crisis situation occur-
ing in the community involving
welfare activities should be made A D A D A D A D
the direct concern of the
council.
6. The council should not hesi
tate to voice its opinion and
take direct action on matters of A D A D A D A D
community improvement even if
they are highly controversial.
7. The council's main focus
should be the development and
coordination of a community-wide
program of case work services.
This entails planning for im- A D A D A D A D
provement and possible exten
sion of existing ones and con
sideration and possible initi
ation of new services.
8. The council may, if it
chooses, raise money for a
cause or program which it con
siders in line with its goals.
9. Council members should be
chosen on the basis of their
generally recognized position in
the community so that the coun- A D A D A D A D
cil's prestige and chances of
getting its recommendations
adopted are thereby enhanced.
10. The need for community
councils such as ours stems
primarily from the need to keep
social agencies in line, to re
duce competition between them, A D A D A D A D
and to see to it that they live
up to the expectations of the
taxpayer and the Community
Chest contributor.
11. The real effectiveness of
the council should be measured
in terms of whether its members
211
Worker LaYman Staff My Own
are growing in their ability to
identify problems, skill in goal
setting and the use of group A D A D A D A D
processes in decision making and
not so much the quality of its
immediate plans and projects.
12. Altho the group process is
used in decision making in the
council it is a function of the
leadership to see that this A D A D A D A D
group process yields the results
which the experts in welfare
planning have in mind.
13. The council should serve as
a sounding-board where persons
or groups who are promoting A D A D A D A D
plans or programs for community
improvement would be welcomed.
14. The need for community
councils such as ours stems pri
marily from the complexity and
specialized nature of special A D A D A D A D
agencies and programs and the
need for continuous and system
atic coordination of such.
C. CLARITY OF ROLE EXPECTATIONS
The following asks that you attempt to estimate the
clarity and definiteness of role behavior expectations of
council participants in their respective roles. For ex
ample, do you think most laymen usually have a clear pic
ture in mind of what is expected of them as laymen in the
council? - and how about social workers, and about staff?
Directions:
TndTcate your view of each statement below by plac
ing a check mark in one of the four squares which in your
judgment best describes the prevailing situation.
212
Usually
Clear
More
T imes
Clear
Than
Unclear
More
T imes
Unclear
Than
Clear
Usually
Unclear
1. The way I view my own
role.
2. The way I think most
LAYMEN in the council view
their own role.
3. The way I think most
SOCIAL WORKERS in the coun
cil view their own role.
4. The way I think STAFF
views its own role in the
council.
D. "WHO SAID THAT"
The statements below, altho not actual quotations
from council meetings might well have been made in council
meetings. Of whom, in your judgement, would they be most
typical or characteristic? - social workers in the council,
laymen in the council, or staff?
Directions: After reading a statement ask yourself,
"of all the persons performing roles in the council, which
one would be most likely to have made such a statement?"
If, for example",' it be most typical of a social worker,
draw a circle around Sw after the given statement; if most
typical of a layman on the council, circle L; if it sounds
like a staff member talking, circle St. If, however, you
view the statement as not necessarily characteristic of any
of the above then place a check mark in the None column.
Caution: Don’t try to take individual exceptions
into account. Your first reaction is what we want.
None
1. "The situation in the community is
deplorable and something should be done Sw L St
about it immediately."
2. "I think we should postpone action on
the matter until we have had time to get Sw L St
more facts."
213
None
3. "What is Planning and Research Council
likely to do with such a recommendation?"'
Sw L St
4. "Isn’t anyone going to defend the pro
fessional social worker’s point-of-view?"
Sw L St
5. "Are the general interests of the com
munity being best served by such an action?
, i Sw L St
6. "Would someone please inform me as to
what this is all about?"
Sw L St
7. "Isn’t this more properly a function of
the private agencies?"
Sw L St
8. "May I remind you that one of the pur
poses of the council is to discover unmet
needs in the community?"
Sw L St
9. "What do we hope to achieve by setting
up such a committee?"
Sw L St
10. "As your vice-chairman I call the
meeting to order."
Sw L St
11. "I think Staff should remain neutral
in such matters."
Sw L St
12. "Which agency is most able to take on
such a function if we should decide to
recommend it?"
Sw L St
13. "What do the laymen of the council
think of the recommendation?"
Sw L St
E. ROLE RESPONSIBILITY
While all members of the council are f ree t ci initi-
ate action or "carry the bail" so to speak, on any matter,
we would like to know to which of the three roles you
would attach the ma.jor responsibility for the following
named tasks. Which role should ordinarily take the lead?
Likewise, which of the three roles would you view as having
least responsibility in this regard.
Directions: If you think the social worker, for
example, should assume the most responsibility for "carry-
214
ing the ball" on a given item then draw a circle around Sw
in the "Most" column. Then indicate which role should be
held least responsible for taking the lead and circle the
appropriate symbol in the "Least" column. (L stands for
layman; St stands for staff.)
Most Least
1. Preparation of agenda for regular
council meetings.
Sw L St Sw L St
2. Keeping Boarding Home rates in line
with rising costs.
Sw L St Sw L St
3. Reconciling differences of opinion
that have arisen between social workers.
Sw L St Sw L St
4. Meeting with the city mayor when
council concern touches on City Council
jurisdictions and interests.
Sw L St Sw L St
5. Interpretation of council actions to
citizen groups.
Sw L St Sw L St
6. Bringing up welfare matters for coun
cil consideration that have or may be
come politically controversial.
Sw L St Sw L St
7. Work out inter-agency agreements
where differences of opinion and hard
feelings may have arisen.
Sw L St Sw L St
8. The increase of services to old peo
ple in the community who are not receiv
ing old age assistance.
Sw L St Sw L St
9. Proposal of the transfer of services
being performed by one agency to some
other agency.
Sw L St Sw L St
10. Attempt to work out differences of
opinion which may have developed between
members of a special study committee.
Sw L St Sw L St
11. Increase the number of Day Care
Homes for children and counselling ser
vices for working mothers.
Sw L St Sw L St
12. Help newly elected laymen to become
oriented.
Sw L St Sw L St
215
Most Least
13. Council goal setting. Sw L St Sw L St
14. Seeing to it that the professional
language used by the social worker is
made understandable to the layman on
council.
Sw L St Sw L St
15. Reduction of unnecessary duplication
overlapping
of services.
Sw L St Sw L St
16. Maintain a balanced program of wel
fare services for the total community.
Sw L St Sw L St
17. Interpretation of staff ideas and
plans to council members.
Sw L St Sw L St
F. INTER-ROLE RELATIONSHIPS
No doubt but what there are many reasons why per
sons, such as yourself, spend time in community activities
such as this. Some possible ones are listed below. We
would like your response to each one listed.
Directions: Please indicate, after each item,
whether you would consider the given item a major factor of
consideration in your willingness to spend time in the
council activities; whether it is a factor of some consid
eration but short of being major; or whether it is of no
consideration whatsoever. A check mark in the appropriate
square is sufficient.
Of MAJOR Of SOME Of NO
P . Consider- Consider- Consider-
* * ation ation ation
a. Some kind of civic activ
ity is expected of one in my
kind of work.
b. Puts me in contact with
important people in the com
munity.
c. Personal satisfaction
derived.
216
Of MAJOR Of SOME Of NO
Consider- Consider- Consider
ation ation ation
d. Makes me feel like a
really useful citizen in the
community.
e. This kind of activity is
common practice among persons
with whom I associate.
f. Is potentially helpful in
my business or to my profes
sional advancement.
g. This type of activity is
expected of the spouse of a
person who is or who wishes
to be prominent in the com
munity.
h. Makes constructive use
of my leisure time.
F.2-- Taking into account all community activities in
which you have ever participated please name the one that
has:
a. Interested you the most
b. Been most satisfying to you
Now, please rate your council role in terms of both "inter
est" and "satisfaction" by using your most interesting and
most satisfying community activity (a bove) as the criteria
or point of reference. For example, if your council role
is more satisfying than any other in which you have parti
cipated then check (below) "Interest" in the "More Than"
column. If your council role is less interesting than, for
example your church activities, but still not the most
uninteresting of your activities then check "Interest" in
the "Somewhat Less" column, etc.
217
AK.„, .. Somewhat So much
More than less less there
the best the best th*n ?he is n?
best comparison
a. My INTEREST IN
COUNCIL ACTIVITIES
IS:
b. The amount of
PERSONAL SATISFAC
TION I receive from
council activities
is:
F.3— Most of us are aware that we are influenced in our
views, wherever we function, by the kinds of social con
tacts and experiences we have. Here we are asking you to
try to trace some source of major influence on your behav
ior and thinking as your participate in council affairs.
Directions: Select from the following possible
sources what you consider to be the three most important
general sources of influence of the role you play in the
council and rank them according to their importance - as
you view them. Indicate the most important one on line 1,
the next most important one on line 2, and the third on
line 3.
(1) ____
(2)
(3)
a. The kind of society in which we live and the
way we are expected to do things in our soci
ety.
b. My professional education and/or training.
c. The way my friends and respected associates
tend to look at such matters.
d. In general the way things are done in my
profession or type of work.
e. My understanding of what the general purpose
and scope of the council is.
f. My observations of the way things are done in
the council.
g. My most valued organizational affiliations.
h. None of the following seem to be as relevant
as: (specify) ...
218
F.4— Using the above statements (a thru g) what would
be your estimate of the major sources of influence gener
ally for the occupants of each of the following roles in
the council? (changing the word "my" to "their.")
Social Worker Layman Staff
1.__/ 2. / 3.__/ 1.__/ 2.__/ 3. / 1. / 2. / 3. /
F.5— The influence of acquaintanceships or contacts made
in one setting are seldom limited to that setting. They
have a way of "splashing over" into other relationships.
Here we are asking you to make some estimate of some of
your council contacts and their significance to some of
your other "valued" roles (outside of the council).
Directions: Think of the names of persons whom you
have met in the council, or in connection with your council
participation, with whom you have had occasion to have some
subsequent relationships outside of the council, (if new on
the council, think in terms of contacts you hope or would
like to make.) Try to classify these into the broad cate
gories below then check in the appropriate box or boxes.
Highly Moderately Perhaps of
valued or valued or some value,
have occa- refer to or Possibly
sion to refer contact them have referred
to or contact occasionally to or con-
them often tacted them
a. Personal
f rie ndship
b. Helpful in
business, occupa
tion, or profes
sion
c. Helpful in get
ting ahead in the
community
d. Other (specify)
219
F.6— Our friends and acquaintances and associates some
times react differently to the activities in which we our
selves engage. This being the case, we may even "go out of
our way" to let some persons know of certain of our activ
ities while other activities remain undisclosed. Here we
are asking you to attempt to classify your friends (along
with acquaintances and associates) into the three following
categories:
A. Those whom you would like to have know about your
membership and activities in the council.
B. Those whom you would prefer not to know about your
membership and activities in the council.
C. Those who, for you, it makes no difference whether
they know or do not know about your membership and
activities in the council.
Directions: After each set of friends listed below
place a check mark in the square which represents your
preference according to the above three categories.
*Note: Circle the asterisk Would Prefer Makes no
if these items do like them they not differ-
not apply to you. to know know ence
a. My own professional asso
ciates or work partners.*
b. Professional associates
or work partners of my
spouse.*
c. My present employer.*
d. An employer whose estab
lishment (company, firm or
agency) I should like to be
asked to join.
e. My circle of friends at
church or synagogue.
f. My close personal
friends.
g. My club or fraternal
friends.
h. My immediate neighbors.
(The ones with whom I have
anything in common.)
220
Would Prefer Makes no
like them they not differ-
to know know ence
i. Persons I most frequently
entertain in my home or with
whom I "go out" most.
j. Groups to which I am as
piring or in which I should
like to be "accepted."
F.7— Since serving on the council have you ever experi
enced any conflict of loyalties between the interests of
the council and some other group in which you have member
ship you value? (check one!
a. Frequently / Occasionally / Seldom /
Never /
b. If this ever happens at all, in which of the above
named groups (in F.6) is this (conflict of loyalty between
council matters and outside interests and friends) most
likely to occur?
c. Any adjustment required on your part as to any dif
ferences between council procedures and your usual mode of
operation? (specify) .
G. ROLE INVOLVEMENT
Most of us are unable or unwilling to become "in
volved" in all of our community roles with the same degree
of intensity. Time and interest and importance are very
real factors. Granting this, we would like some indication
of your degree of "involvement" in your council role inso
far as the following items are able to determine it.
Directions: Please check the following items in the
manner which best describes your situation.
1. During the past year how many of the regular council
meetings were you able to attend? Most of them ___/
About half / Seldom able to attend /
221
2. During the past year have you served on any standing
committees ( ) or on any special committees ( J?
One / Two / More than two /
3. If your answer to No. 2 is "yes:"
a. Approximately how many such meetings did you
attend in all? Less than three / Three to
five __/ More than five /
b. Did you volunteer ( ) or were you asked to
serve ( )?
c. Did you serve the council on any special assign
ments ( ) or delegations, etc. (___ )? If
yes, approximately how much time did you devote
to such?
About the same amount of time one to three
committee meetings would probably take /
Three to five / More than five /
4. Has your work on the council necessitated your dropping
any other activities or kept you from taking on some
activity which you would have otherwise taken on?
Yes / Not necessarily /
5. How would you characterize your participation in the
regular council meetings?
___ a. Frequently involved in discussion or debate.
_____ b. Participate occasionally in discussion or
debate.
_____ c. Ask an occasional question.
_____ d. Listen and vote.
6. Which of the following statements comes closest to
explaining your "kind" of council participation?
_____ a. Often feel that my experience and/or position
justifies my speaking up.
_____ b. Occasionally feel that my position and/or ex
perience is relevant to the issue at hand.
_____ c. Often feel I should speak up bimt am hesitant
about parading my opinions.
d. Seldom feel I have anything significant to
contribute.
_____ e. Usually am at a loss to know what is going on.
7. During the past year, have there been any decisions made
or actions taken in the council which disturbed you or
gave you cause to be really "bothered?" Yes /
No /
If1 your answer is "yes" - Which of the following -
statements best characterize your subsequent actions?
222
_____ a. Followed it up by trying to get more informa
tion.
_____ b. Got in touch with womeone whom I thot was in
a position to do something about it.
_____ c. Joined with some others who felt as I did to
try to do something about it.
_____ d. Brought the matter up again for consideration
at a later council meeting.
____ e. Decided that time would take care of the mat
ter - took no other action.
_____ f. Was somewhat upset at first but dismissed it
subsequently.
8. Are welfare matters which are dealt with in the council,
ever brought up for discussion by you in informal conversa
tions or discussions with your friends?
Frequently. / Occasionally / Seldom / Never /
A. If your answer is "frequently" or "occasionally" -
Which of the following best characterizes your
reason for doing so?
_ a. I think it part of my job as a council member
to let them know what is taking place in
the council.
_____ b. My friends are interested in what the council
does.
c. My friends can help me think thru some of the
matters on which I am called upon to vote
in the council.
____ d. Such matters make good conversation.
____ e. I like my friends to know that I am involved
in serious community matters.
_____ f . Other
your answer to No. _8 JLs "seldom" or "never" -
Which best characterizes your reason?
_____ a. Council matters should stay within the coun
cil.
_____ b. My friends wouldn't be interested.
_____ c. My friends wouldn't be impressed with council
matters.
_____ d. When I leave the council I shut such matters
out of my thinking, for the most part, until
next meeting.
_____ e. Other ___________________________________,
9. When referring to some action of the council of which
you approve which set of pronouns (set a or set b) would
you prefer to use? a_____/ (We (They
b / a I Our b (Their
(Us (It
223
When referring to some action of which
you disapprove which set would you prefer
to use? a /
10. Do you ever find yourself in a position of voluntarily
defending the council or its actions (decisions) outside of
the council?
Frequently / Occasionally / Seldom /
Never /
A. If your answer is "frequently1 1 or "occasionally" -
Among which oTT"the foilowing persons or kinds of
groups is this most likely to happen?
_____ a. Professional associates or work partners.
______ b. Circle of personal friends.
______ c. Club or fraternal friends.
_______ d. Persons I most frequently entertain in my
home or with whom I "go out" most.
______ e. Groups to which I am aspiring or in which I
should like to be "accepted."
______ f. The professional associates or work partners
of my spouse.
______ g. My employer or "boss."
. h. My immediate neighbors (the ones with whom I
have something in common).
______ i. Groups with which I should like to be disas
sociated but because of certain "obliga
tions" am unable to do so.
B. If your answer (to No. 10) is "seldom" or "never" -
Among which of the above named persons would you
most likely refrain from defending the council
or its actions? _ __ _______ ________ __
(Thank you very much)
COVER LETTER SENT WITH EACH
QUESTIONNAIRE
October 13, 1958
Dear Fellow Case Work Council Member:
During the years I have served on the St. Paul Case
Work Council I have become curious about the way council
members view their respective roles in the council. (By
"role" I simply mean the part they play in the council.)
When the time came for me to select a Ph.D. dissertation
topic (in social psychology, University of Southern Califor
nia ) I suggested this and it was accepted by my study com
mittee . Subsequently I submitted the project to both the
Case Work Council and the Planning and Research Council and
was given the "green light" by both.
As a sociologist I am interested in examining group
structures within which on-going social processes take
place, such as we have here in our council. Some of this
"structure" is expressed formally in by-laws and the like.
But much of it resides in the perceptions which members
have about their own roles and the roles of others in the
council. This is my interest.
There appear to be three main role participants:
layman, social worker (both agency representatives), and
staff (chest and council personnel). I want to get at the
amount of consensus which prevails among members concerning
the behavior expectations which each has for these roles,
something of the satisfactions which come to the role in
cumbents, degrees of involvement in the role, and the im
pact of one’s council role on some out-of-council roles.
The enclosed questionnaire, which I am asking you
to fill out, is quite lengthy I must admit. Unfortunately,
this kind of research has a way of imposing on the time of
many. But I know of no other way of doing it; and I think
you will agree with me that it is worth doing. You see, my
study represents a rather intensive inquiry into one coun
cil rather than an extensive coverage of many councils.
The questionnaire does not deal in "right" and "wrong"
answers. It asks simply for the way you "look at" these
224
225
roles and the way you think others "look at" them.
May I assure you of complete confidentiality of your
responses I True, I have coded the questionnaire before
sending it to you. But no one but myself will know the
identity of the respondents. The reason why I want to know
the identify is this: After compiling the information from
the questionnaires I expect to follow-up with interviews
of a random sample of council members. I wish to pair the
interview data with the questionnaire findings in the final
presentation. This, I hope, will help overcome some of
the objections often raised about questionnaire studies.
But remember, neither your identity nor that of the
agency you represent will be revealed in the dissertation
or in any subsequent use of the findings.
Please be assured that your help in this project
will contribute to the growing reputation which our commun
ity has throughout the country for its explorations into
the frontiers of human behavior. And for your generous ef
forts in behalf of my personal academic goal I shall be
most grateful.
Thankyou sincerely,
(signed) Paul M. Berry
APPENDIX B
ORGANIZATIONAL CHART OF THE
GREATER ST. PAUL COMMUNITY CHEST
AND COUNCILS
MEMBER AGENCIES AND ORGANIZATIONS
OF THE SAINT PAUL CASE WORK OOUNCIL
GREATER ST. PAUL COMMUNITY CHEST AND COUNCIL, INC.
ORGANIZATIONAL CHART
A pril 30, 1949
CORPORATE MEMBERSHIP
C o n trib u tin g
Honorary
Organi z a tio n a l
PARTICIPATING
AFF IL I ATEO
______________ BOARD OF DIRECTORS________________
Publ i c at L a rg e ________________________________ 9
Campaign Com m ittee____________________________ 3
Budget Com m ittee_______________________________ 3
Planning and Research C o u n c il_______________ 9
Chm. o f Coordinating C o u n c ils _______________ 3
Chm. o f Planning & Research C o u n c il I
28
E x -O ffic io : Chm o f Budget, P u b lic ity ,
Campaign & Nominating Committees________
Casework
Health
Lei sure Time
STAFF
~n r ~
Budgeting
Audi t i ng
Col 1e c ti ng
Commun i ty
Inform at ion
EXECUTI VE
D I RECTO R
STAFF
Publi c
R elatio n s
1
Special G ifts
Employees Div.
Neighborhood Div.
CAM PA IG N
COMM ITT EE
Chm., V. Chm.
Appointed by
Board
BUDGET
COMMITTEE
Chm., V. Chm.
Appt. by Board
12 Elected by
Plan. & Res. Cl.
15 Elected by
Board
PLANNING & RESEARCH COUNCIL
Chairman — Appointed by
Chairman — Coordinating
Vice Chm.— Coordinating
Members a t Large Elected
Large Elected Members at
Presi dent,
B oard_____________|
Councils ________ 3
Counci 1 s _________9
by Board ________ 3
by Above | 6 _____ 6
Chm. o f Budget,
Campaign and P u b lic ity
C o m m itte e s _____________________ 4
26
RESEARCH &
ST A T IS T IC S
P r io r i ty
Pol i cy
P roj ect
PU B L I C
RELATI ONS
Chm., V. Chm.
Appointed by
Board
Admi n i st ra t ion
by W ild er C h a rity
C 0 0 R DI NA TI NG COUNCI L S
ST. PAUL AREA
PUBLIC HEALTH CL.
Same as For
Case Work
Counci 1
GREATER ST. PAUL
CASE W ORK COUHCIL
EXECUTIVE COMM.
Chm. (Layman)
3 Vice Chm.
1 CHEST AGENCY
1 NON-CHEST
1 PUBLIC AGENCY
3 Members at Large
2 Del egates from each
Organizational Mem.
LEISURE TIME
ACT !V IT IE S COUNCIL
Same as For
Case Work
Counci 1
MEMBER AGENCIES AND ORGANIZATIONS OF THE
SAINT PAUL CASE WORK COUNCIL
American National Red Cross, St. Paul Chapter
Board of Christian Service (Lutheran Home for Invalids)
Boy Scouts of America - St. Paul Area Council
Bureau of Catholic Charities of St. Paul, Inc.
Capitol Community Services, Inc.
Catholic Infant Home
Child Guidance Clinic
Children's Home Society of Minnesota (adoption)
Children's Service, Inc. (psychiatric treatment)
Christ Child Community Center
City and County Detention and Corrections Authority
Civil Defense Office
College of St. Catherine (pre-professional social work
training )
Community Workshop (rehabilitation)
Council of Jewish Women
County Welfare Board of the City of St. Paul and Ramsey
County
Crispus Attucks Home (home for aged)
Family Service of St. Paul
Fourth District Minnesota Nurses Association
Goodwill Industries, Inc.
Hallie Q. Brown Community House
Hamline University (pre-professional social work training)
Hamm Memorial Psychiatric Clinic
Housing and Redevelopment Authority of the City of St. Paul
Inter-Club Council
International Institute
Jaycee Wives Club
Jewish Community Center
Jewish Family Service
Jewish Home for the Aged of the Northwest
Jewish Vocational Service
Junior League of St. Paul, Inc.
Luthern Committee for Religious Social Work
Lyngblomsten Home for the Aged
Macalester College (pre-professional social work training)
Merriam Park Community Center
Minnesota Department of Public Welfare
Minnesota Prisoner's Aid Society
Minnesota State Youth Conservation Commission
Neighborhood House Association, Inc.
Nursery School for Mentally Retarded Children
228
229
Ramsey County Home School for Boys
Ramsey County Home School for Girls
Ramsey County Juvenile Court
Ramsey County Probation Department
Saint Joseph’s Home for Children
St. Paul Association for Retarded Children
St. Paul C.I.0.Council
St. Paul Council of Churches
St. Paul Council of Parent Teachers Association
St. Paul Junior Chamber of Commerce
St. Paul Public Schools (Division of Special Services, e.g.,
school social workers)
St. Paul Rehabilitation Center
St. Paul Society for the Blind
St. Paul Urban League
Salvation Army, The
Travelers Aid
Twin City Regional Hospital Council
Twin City Linnea Society
University of Minnesota (School of Social Work)
Veterans Administration Center
Veterans Administration Hospital
Volunteer Bureau, Inc.
Wilder Day Nurseries
Wilder Dispensary
Woman's Auxiliary to Ramsey County Medical Society
Young Men’s Christian Association
Young Women’s Christian Association
APPENDIX C
STATISTICAL TABLES
REFERRED TO IN THE DISSERTATION
231
TABLE 1-C
PERCENTAGE AND NUMERICAL DISTRIBUTION
OF STATED ROLE CLARITY
WITH WHICH EACH OUNCIL ROLE IS PERCEIVED3
BY ALL DELEGATES, BY LENGTH OF COUNCIL TENURE
Role
Be ing
Per
ceived
N
0 /
/ o N
c : /
/J N
f ' /
/O N
%
2- 13
37. 19 54. 3
9. 35 51.
Soc ial 3+ 16
49. 12 36. 5 15. 33 49.
Work
T otal 29 43. 31 45. 8 12. 0^
00
o
100.
2- 2 6. 15 53. 18 51. 35 51.
34 5 15. 15 46. 13
39.
33
49.
Layman
Total 7 10. 30 44. 31 46.
u
00
vO
100.
2- 30 33. 6 17. 0 0. 36 51.
3+ 24 70. 8 24. 2 5. 34 49.
Staff
Total 54 77. 14 20. 2 3. 70 100.
2- 12 33. 10 28. 14 39. 36 51.
34
14 41. 12 35. 8 24. 34 49.
Own
T otal 26 37. 22 31. 22 31. 70 100.
aSee footnote page 130 on "indirect1' perceptions as
symbolized by REOP.
2- means short-term participants (two years or
less); 3+ means long-term participants (three years or
more ).
cReduced N due to two incomplete questionnaires.
All
Stated Clarity
of Role Perception
Delegates More More
^en9Th Usually Clear Unclear T ,
Of Council clear Than Than Total
Tenure0 Unclear Clear
23 2
TABLE 2-C
PERCENTAGE AND NUMERICAL DISTRIBUTION
OF STATED ROLE CLARITY
WITH 'WHICH EACH COUNCIL ROLE IS PERCEIVED
BY EACH OF THE ROLE JUDGE GROUPINGS
Role
Being
Pe r-
ceived
Role
Judge
Grouping
Stated Clarity
of Role Perception
More More
Usually Clear Unclear
Clear Than Than
Unclear Clear
N % N % N %
Total
N %
MSW 8 33. 11 46. 5 21. 24 35.
Non-MSW 10 48. 9 43. 2 09. 21 31.
Social
Work
Lay 11 48. 11 48. 1 04. 23 34.
Total 29 43. 31 45. 8 12. 68a 100.
MSW 2 08. 11 46. 11 46. 24 35.
Non-MSW 2 10. 12 57. 7 33 . 21 31.
Layman Lay 3 13. 7 30. 13 57. 23 34.
T otal 7 10. 30 44. 31 46. 68a 100.
MSW 17 31. 7 50. 1 50. 25 3 36.
Non-MSW 16 30. 5 36. 1 50. 22 31.
Staff Lay 21 39. 2 14. 0 00. 23 33.
Total 54 77. 14 20. 2 03. 70 100.
MSW 8 31. 11 50. 6 27. 25 36.
Non-MSW 10 38. 5 23. 7 32. 22 31.
Own Lay 8 31. 6 27. 9 41. 23 33.
Total 26 38. 22 31. 22 31. 70 100.
aReduced N due to 2 incomplete questionnaires.
233
TABLE 3-C
CHI SQUARE TABLE SHOWING THE EXTENT
OF INDEPENDENCE OF STATED ROLE CLARITY
AND LENGTH OF COUNCIL TENURE,
FOR EACH COUNCIL ROLE
Role
Being
Per
ce ived
Length
of
Council
Tenure3
Stated Clarity
of Role Perception
Usually
Clear
More
Clear
Tha n
Unclear
More
Unclear
Than
Clear
Total
Social
'No r k
2-
3+
Total
13
16
29
19
12
31
3
5
8
35
33
68b
x2 = 2.27
P> .3 0 < .50
2- 2 15 18 35
Lay
3-*
5 15 13 33
Total 7 30 31 68b
x2 = 2.00
P > .30 < .50
Own
2-
3h
T otal
12
14
26
10
12
22
14
8
22
36
34
70
X2 » 1.94
P > .30 < .50
a2- means short-term participants (2 years or less);
3*. means long-term participants (3 years or more).
^Reduced N due to two uncompleted questionnaires.
■' 1 ' i " ■ !. f f - e j — - m - ............* - " - — i a i l . .
234
TABLE 4-C
CHI SQUARE TABLE SHOWING THE EXTENT
OF INDEPENDENCE OF STATED ROLE CLARITY
AND TYPE OF ROLE JUDGE GROUPING,
FOR EACH COUNCIL ROLE
Role
Being
Per
ce ived
of
Stated Clarity
Role Perception
Role
Judge
Grouping
Usually
Clear
More
Clear
Than
Unclear
More
Unclear
Than
Clear
Total
Social
Work
Social
Worker
Laymen
18
11
20
11
7
1
46
23
T ota 1 29 31 8 68
x2
-1.87
P > ,30< .50
Lay
Social
Worker
Laymen
0a
0a
27
10
18
13
45
23
T otal 0 37 31 68
X2 - 1.64
P >
.20
Own
Social
Worker
Laymen
18
8
16
6
13
9
47
23
Total 26 22 22 70
x^ a 1. 01
P>
.50 < .70
aCell size required combining with "more clear than
unclear."
235
TABLE 5-C
CHI SQUARE TABLE SHOWING THE EXTENT
OF INDEPENDENCE OF ROLE ANCHORAGE
IN PROFESSIONAL ORIENTATION
AND ROLE JUDGE GROUPING
Role
Judge
Helpful in Business, Occupation,
or Profession
Total
Grouping
Highly Moderately Possibly
Social
Worker
24 16 6 46
Layman 3 6 7 16
T otal 27 22 13 62
x2
p >
a 8.27
,10< .02
TABLE 6-C
CHI SQUARE TABLE SHOWING THE EXTENT
OF INDEPENDENCE OF ROLE ANCHORAGE
IN PROFESSIONAL ORIENTATION
AND LENGTH OF COUNCIL TENURE
Length
of
Helpful in Business, Occupation,
or Profession
Total
vOuncx x
Tenure3 Highly Moderately Possibly
2- 15 12 6 33
3+
12 10 7 29
Total 27 22 13 62
x2
P >
=- .32
.80 <. .90
a2- means short-term participants (two years or
less); 3^. means long-term participants (three years or more)
236
TABLE 7-C
CHI SQUARE TABLE SHOWING THE EXTENT
OF INDEPENDENCE OF ROLE INVOLVEMENT
AND ROLE CONSENSUS,
FOR EACH CATEGORICAL ROLE
Role
Being
Perceived
Role
Consensus
Score
Role Involvement Scores
T otal
0 - 12 13 - 24 25 & up
0 - 2 8 10 18 6 34
Social
Work
29 & up
5 22 9 36
Total 15 40 15 70
x2 =
P> •
2.59
20 C .30
0 - 2 2 9 17 10 36
Lay 23 & up 6 23 5 34
Total 15 40 15 70
x^ a
P> .
3. 16
20 < .30
0- 2 6 8 20 7 35
Staff 27 & up 7 20 8 35
T otal 15 40 15 70
• .13
P > .90< .95
23 7
TABLE 8-C
ail SQUARE TABLE SHOWING THE EXTENT
OF INDEPENDENCE OF ROLE INVOLVEMENT
AND STATED COUNCIL ROLE SATISFACTION
FOR ALL COUNCIL DELEGATES
Extent of Comparative Role
Satisfaction
Involvement
Scores
Same-as or
more-than
the best
Less than
the best
T ota 1
0-17 6 29 35
18 & up 11 22 33
Total 17 51 68a
x2 = 2.45
P> . 10< .20
238
TABLE 9-C
CHI SQUARE TABLE SHOWING THE EXTENT
OF INDEPENDENCE OF ROLE INVOLVEMENT
AND STATED COUNCIL ROLE INTEREST
FOR ALL COUNCIL DELEGATES
Extent of Comparative Role
Interest
8-0 le
Involvement - „ _„
q Same-as or
r0vJ more than Less than Total
the best the best
0
1
■v!
10 20 30
18 & up 14 16 30
Total 24 36 60a
x2 - 1.11
P> .20< .30
aR.educed N due to incomplete questionnaire responses.
239
TABLE 10-C
CHI SQUARE TABLE SHOWING THE EXTENT
OF INDEPENDENCE OF STATED ROLE CLARITY
AND ROLE CONSENSUS
WHEN THE CATEGORICAL ROLES
ARE PERCEIVED3 BY ALL DELEGATES
Categorical
Role Being
Perceived
Ro le
Stated Clarity
of Role Perception
Consensus
Score
Usually
Clear
More
Clear
Than
Unclear
More
Unclear
Than
Clear
Total
0-28 17 11 3 31
Social
Work
29 & up 14 18 5 37
Total 31 29 8 68b
x2 i
P>
n 1 95
.30< .50
0 - 2 2 3 17 15 34
Lay 23 & up 4 14 15 33
T otal 7 31
29 67b
-
x2 ,
P >
= 4Q
.70\ .80
aTest
distribution.
not applied to Staff role because of
See p. 162.
skewed
^Reduced N caused
sponses.
by incomplete questionnaire re-

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Familism, Suburbanization, And Residential Mobility In A Metropolis

PDF

The Construction And Empirical Test Of A Theory Based On Selected Variables In Small-Group Interaction

PDF

An Empirical Examination Of The Relationship Of Vertical Occupational Mobility And Horizontal Residential Mobility

PDF

Interpersonal Relations In Ethnically Mixed Small Work Groups

PDF

Some Social Factors Affecting The Power Structure And Status Of A Professional Association In Reference To Social Work

PDF

Social Class Membership And Ethnic Prejudice In Cedar City

PDF

The Career Business Executive As A Definitive Occupational Type

PDF

Normative values of selected law enforcement officers and adult male offenders

PDF

Reference Group Theory, Selection, And The Images Of Professions

PDF

Social Changes In Selected Institutions Of The Ussr With Special Reference To The Family

PDF

Perception Of The Power Structure By Social Class In A California Community

PDF

Sex-Role Preferences Of Early Adolescents In Relation To Adjustment

PDF

A Critique Of The Concept Ethnocentrism As Set Forth In Selected Social Science Literature

PDF

Middle-Class Marital Roles - Ideal And Perceived In Relation To Adjustment In Marriage

PDF

Selected Social Psychological Factors Related To Viewers Of Television Programs

PDF

An Empirical Study On The Differential Influence Of Self- Concept On The Professional Behavior Of Marriage Counselors

PDF

Role Expectations Of American Undergraduate College Women In A Western Coeducational Institution

PDF

Social Components Of Housing Cost In The Western Metropolis

PDF

A Study Of Relationships Between Occupational And Marital Roles And Marital Adjustment

PDF

The Effect Of Differential Treatment On Attitudes, Personality Traits, And Behavior Of Adult Parolees

Asset Metadata

Creator Berry, Paul Mccoy (author)

Core Title Consensus Of Role Perceptions In A Welfare Planning Council

Degree Doctor of Philosophy

Degree Program Sociology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,psychology, social

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Neumeyer, Martin H. (committee chair), McDonagh, Edward C. (committee member), Stinson, Malcolm B. (committee member), Van Arsdol, Maurice D., Jr. (committee member), Vincent, Melvin J. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-54105

Unique identifier UC11357586

Identifier 6000555.pdf (filename),usctheses-c18-54105 (legacy record id)

Legacy Identifier 6000555.pdf

Dmrecord 54105

Document Type Dissertation

Rights Berry, Paul Mccoy

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA